Prognostic Findings for HER2 in Breast Cancer not Reproducible

That HER2 is prognostic of outcome in breast cancer is unquestioned. As Jeffrey Ross at Albany Medical College put it: “Today, no one I know doubts in any way that, in the absence of anti-HER2 therapy, HER2+ breast cancer is an unfavorable subtype and HER2+ status by IHC or FISH is a significant and independent prognostic factor.”

Ross helped shape HER2’s reputation as a particularly aggressive form of breast cancer. In 1998, Ross and co-author Jonathan Fletcher published a review of 47 studies of HER2. Each study was checked for an “impact” on prognosis, either univariate or multivariate. (Appendix C lists the 47 studies.)

Univariate findings can be misleading, often losing significance when multiple factors are taken into account. Regarding the more robust multivariate analyses, Ross and Fletcher reported that 28 (60%) of 47 studies found multivariate impact. The remaining 40% of studies either found no multivariate impact or didn’t conduct a multivariate analysis.

Counted by cases, 10,142 (67%) patients out of 15,248 were in studies found by Ross and Fletcher to have a multivariate impact. Their review concludes: “The preponderance of evidence indicates that HER- 2/neu gene amplification and protein overexpression are associated with an adverse outcome in breast cancer.”

However, the review’s conclusion depends on miscategorizing 9 of the 47 papers examined. Correctly categorizing these 9 studies to reflect their actual findings overturns the conclusion that HER2 is prognostic. The preponderance of evidence is inverted and points to no adverse outcome from HER2 (Table 1). Similarly, the number of cases supporting a prognostic value for HER2 fall from two thirds to less than half (Table 2).

Table 1: Number of studies finding HER2 independently prognostic in multivariate analysis

Table 2: Number of cases in studies finding HER2 independently prognostic in multivariate analysis

Appendix A lists the 9 studies and justification for each recoding.

Ross did not dispute the recodings. Provided with the information in Appendix A and asked if he agreed with the recoding, Ross replied: “I am traveling in Europe and have limited time to review. It is certainly possible that the studies you have cited were not perfectly listed in my manuscript from so many years ago.”

Ross and Fletcher’s review suffers from multiple shortcomings. (Appendix B enumerates important but secondary flaws.) However, the miscoding of papers in Ross and Fletcher’s review is sufficient to overturn the paper’s conclusion.

Conflicts of interest

Investigations of HER2 as a prognostic factor produced contradictory findings and argument—resolved by Ross and Fletcher. Of note, commercial interests played a role in several of the studies they reviewed and the review itself.

Among the 47 papers examined, four [4, 10, 29, 48] list at least one author with a corporate rather than academic affiliation. One abstract [49] includes an author who was then a director of diagnostics at Oncor, maker of a HER2 test. All five studies reported HER2 as prognostic.

In their review, Ross and Fletcher report being consultants for Oncor. However, according to Bloomberg, Ross was Medical Director at Oncor beginning in late 1995 and later Chief Medical Officer when his review with Fletcher was published in 1998. Ross confirmed the accuracy of Bloomberg’s information. The FDA rejected Oncor’s test in 1995 but, as reported in the New York Times, Oncor won approval in 1998.

Conclusion

It is likely true, as Ross stated, that today no one questions that HER2 is prognostic in breast cancer. However, this supreme confidence needs to be recalibrated.

Appendix A: Recoded papers

Of the 47 studies, the nine below were recoded:

[11] (Battifora et al.): Yes to No

The paper reports: "Stepwise Cox Regression: This analysis identified independent prognostic factors of DFS and OS when all variables were considered together. Independent predictors of DFS included stage of disease, histology, and nuclear grade. Nuclear grade and stage were the only significant predictors of OS."

[14] (Lovekin et al.): Yes to No

The paper reports: “Multivariate analysis (Cox, 1972) was used to identify whether c-erbB-2 was of independent prognostic significance. In the context of the temporal variables, tumour size and lymph node stage, cell membrane staining was found to have independent significance as a prognostic factor but significance was lost when histological grade was included in the analysis."

[16] (Dykins et al.): Yes to NA

No multivariate analysis

[20] (Paterson et al.): Yes to No

The paper does not state HER2 is independently prognostic in a multivariate analysis or provide the statistics relevant to such a statement. The authors do suggest possible confounding of prognostic factors: “our study design precluded direct determination of the interrelationships of c-erbB-2 [HER2] amplification with conventional disease parameters.”

[22] (Molina et al.): Yes to NA

No multivariate analysis

[29] (Press et al.): Yes to NA

No multivariate analysis

[31] (Descotes et al.): Yes to NA

As its title states, the paper is a “correlation study between Her-2/neu amplification and prognostic factors.” No disease outcome data are included in the paper.

[34] (Têtu et al.): Yes to No

The paper reports that HER2 was predictive of treatment resistance, not prognostic: “The difference in survival rates between cases was only significant among patients submitted to adjuvant chemotherapy or hormone therapy."

[47] (Charpin et al.): Yes to NA

No multivariate analysis

Appendix B: Additional methodology issues

Inclusion criteria

How the 47 papers reviewed by Ross and Fletcher were selected is not described. In email, Ross wrote that “if you just limit the publications cited to those finding HER2 positive rates between 10 and 30% the prognostic impact of HER2+ status in the pre-anti-HER2 targeted therapy era was profound.”

However the review includes Dittadi et al. [44] which describes a “high risk” group comprising 44% of all cases, well above 30%. Ross and Fletcher count the study as supporting the independent, multivariate prognostic impact of HER2.

Berger et al. [5] and Descotes et al. [31] only examine correlations between biomarkers not with disease outcomes and should not have been included.

Ross and Fletcher included two studies [42, 49] for which there are only abstracts. More generally, the studies included were not graded for quality.

An unknown number of papers were omitted, potentially introducing a selection bias. An omitted paper from Zhou et al. (1989), for example, found no prognostic value for HER2. On the other hand, Wright et al.  (1989) also was not included but found HER2 independently prognostic. Other possible biases in the literature, against publishing, for example, are not examined.

Reviews frequently require a minimum number of cases for a study to be included. Indeed, a number of the papers reviewed by Ross and Fletcher attribute the conflicting results in HER2 studies in part to studies with small numbers of cases.

One study [43] had 37 cases. Ross and Fletcher record it as finding HER2 prognostic in univariate analysis but the paper contains no p values, perhaps because n is so small. O’Malley et al. [41] does not state the number of HER2 positive cases that provided the basis for the conclusion that HER2 was prognostic in multivariate regressions. (The corresponding author did not reply to an email inquiry.)

A 2002 review of prognostic factors in node-negative breast cancer specified inclusion criteria and set a minimum number of cases (200). The paper concluded HER2 is not prognostic.

No quantification of prognostic influence

Ross and Fletcher do not provide summary statistics based on a pooling of results. Heterogeneity of the study designs perhaps made this difficult or impossible. However, if heterogeneity prevented statistical summarization, that would be an important finding to report.

The review includes a table of 18 prognostic factors in breast cancer but makes no comparison of their relative strength and clinical value. The prognostic value of HER2 varied widely. In [18], the p value rested at precisely 0.05. Nodal status and tumor size were vastly more prognostic: p < 0.0001 and p = 0.003 respectively. Quénel et al. [39] found HER2 weakly prognostic: "in our hands, c-erbB2 [HER2] had a poor prognostic value in comparison with the classical prognostic variables…” However, whether such weak prognostic value is general among the papers finding HER2 prognostic is not examined by Ross and Fletcher.

Differences in treatment of cases occur within and between studies but the paper does not control for confounding of prognosis with predicting resistance to treatment.

HER2 positive undefined

Different studies used different definitions of HER2 positive. Even today, the definition of HER2 positive and the search for the best HER2 assay continue to be active areas of study. Ross and Fletcher identify the different assays used in HER2 determination (e.g. IHC, FISH) but cut points are not extracted.

Three studies [4, 20, 30] found amplification of HER2 prognostic. But each used a different cut off for gene copy number: six, three, and seven respectively. A single threshold would likely change the findings of these studies and affect the count of studies finding HER2 prognostic.

Some papers determined cutoffs and comparison groups based on achieving statistical significance. One study [44] found HER2 prognostic by creating a “high risk” group that combined cases with the lowest and the highest expression of p185. The low expression group had the worst outcome. Dittadi et al. go on to conclude p185 was independently prognostic in a multivariate analysis. Slamon et al. [4] simply dropped 23 cases with 2-5 copies of HER2. This remarkably unscientific omission enabled comparing a group with one copy of HER2 to those with six or more, providing the basis for the claim HER2 was independently prognostic in a multivariate analysis.

Negative findings not counted, contradictory findings are

Studies with even a single positive finding were counted by Ross and Fletcher as evidence supporting HER2 as a prognostic factor. The number of negative findings is not reported. For example, O’Reilly et al. [19] found HER2 prognostic for relapse-free survival but not overall survival in node-positive disease. Ross and Fletcher count [19] as one of 28 papers supporting the finding that HER2 is prognostic.

Quénel et al. [39], conducted multivariate analyses for three clinical outcomes for three groups. Among the nine tests in total, HER2 showed prognostic value in two and no prognostic value in seven. Ross and Fletcher count [39] among the papers showing that HER2 is prognostic.

Ross and Fletcher’s design also allows studies with opposing findings to be counted as finding HER2 prognostic.  For example Gusterson et al.  [27] found HER2 prognostic in node-positive but not node-negatives patients. However, Giai et al. [32] found the opposite. The papers contradict each other but both are counted as showing HER2 is prognostic by Ross and Fletcher.

Appendix C: Citations for the 47 studies reviewed by Ross and Fletcher

Footnote numbers are those used by Ross and Fletcher. The same numbers are used throughout this article.

4 Slamon DJ, Clark GM, Wong SG et al. Human breast cancer: correlation of relapse and survival with amplification of the Her-2/neu oncogene. Science 1987;235:177-182.

5 Berger MS, Locher GW, Saurer S et al. Correlation of c-erb B2 gene amplification and protein expression in human breast carcinoma with nodal status and nuclear grading. Cancer Res 1988;48:1238-1243.

6 van de Vivjer MJ, Peterse JL, Mooi WJ et al. Neu-protein overexpression in breast cancer. N Engl J Med 1988;319:1239-1245.

7 Heintz NH, Leslie KO, Rogers LA et al. Amplification of the c-erb B-2 oncogene in prognosis of breast adenocarcinoma. Arch Pathol Lab Med 1990;114:160-163.

8 Tsuda H, Hirohashi S, Shimosato Y et al. Correlation between histologic grade of malignancy and copy number of c-erbB-2 gene in breast carcinoma. A retrospective analysis of 176 cases. Cancer 1990; 65:1794-1800.

9 Borg A, Tandon AK, Sigurdsson H et al. HER-2/neu amplification predicts poor survival in node-positive breast cancer. Cancer Res 1990;50:4332-4337.

10 Paik S, Hazan R, Fisher ER et al. Pathologic findings from the nations’ surgical adjuvant breast and bowel project: prognostic significance of erb B2 protein overexpression in primary breast cancer. J Clin Oncol 1990;8:103-112.

11 Battifora H, Gaffey M, Esteban J et al. Immunohistochemical assay of neu/c-erb B-2 oncogene product in paraffin-embedded tissues in early breast cancer: Retrospective follow-up study of 245 stage I and II cases. Modern Pathol 1991;4:466-474.

12 Kallioniemi OP, Holli K, Visakorpi T et al. Association of Cerb B2 protein over-expression with high rate of cell proliferation, increased risk of visceral metastasis and poor long-term survival in breast cancer. Int J Cancer 1991;49:650-655.

13 Clark GM, McGuire WL. Follow-up study of HER-2/neu amplification in primary breast cancer. Cancer Res 1991;51:944-948.

14 Lovekin C, Ellis IO, Locker A et al. C-erb B2 oncoprotein expression in primary and advanced breast cancer. Br J Cancer 1991;63:439-443.

15 McCann AH, DeDervan TA, O’Regan M et al. Prognostic significance of C-erb B2 and estrogen receptor status in human breast cancer. Cancer Res 1991;51:3296-3303.

16 Dykins R, Corbett IP, Henry J et al. Long term survival in breast cancer related to overexpression of the C-erb B2 oncoprotein: an immunohistochemical study using monoclonal antibody NCL-CB11. J Pathol 1991;163:105-110.

17 Rilke F, Colnaghi MI, Cascinelli N et al. Prognostic significance of HER-2/neu expression in breast cancer and its relationship to other prognostic factors. Int J Cancer 1991;49:44-49.

18 Winstanley J, Cooke T, Murray GD et al. The long term prognostic significance of C-erb B2 in primary breast cancer. Br J Cancer 1991;63:447-450.

19 O’Reilly SM, Barnes DM, Camplejohn RS et al. The relationship between C-erb B2 expression, and s-phase fraction in prognosis in breast cancer. Br J Cancer 1991;63:444-446.

20 Paterson MC, Dietrich KD, Danyluk J et al. Correlation between C-erb B2 amplification and risk of recurrent disease in node-negative breast cancer. Cancer Res 1991;51:556-567.

21 Toikkanen S, Helin H, Isola J et al. Prognostic significance of Her-2 oncoprotein expression in breast cancer: a 30-year follow up. J Clin Oncol 1992;10:1044-1048.

22 Molina R, Ciocca DR, Candon AK et al. Expression of HER-2/neu oncoprotein in breast cancer: a comparison of immunohistochemical and western blot techniques. Anticancer Res 1992;12:1965-1991.

23 Noguchi M, Koyasaki M, Ohta N et al. c-erb B-2 oncoprotein expression versus internal mammary lymph node metastases as additional prognostic factors in patients with axillary lymph node-positive breast cancer. Cancer 1992;69:2953-2960.

24 Allred DC, Clark GM, Tandon AK et al. HER-2/neu nodenegative breast cancer: prognostic significance of overexpression influenced by the presence of in-situ carcinoma. J Clin Oncol 1992;10:599-605.

25 Babiak J, Hugh J, Poppeme S. Significance of c-erb B-2 amplification in DNA aneuploidy. Analysis in 78 patients with node-negative breast cancer. Cancer 1992;70:770-776.

26 Tiwari RK, Borgen PI, Wong GY et al. HER-2/neu amplification and overexpression in primary human breast cancer is associated with early metastasis. Anticancer Res 1992;12:419-426.

27 Gusterson BA, Gelber RD, Goldhirsch A et al. Prognostic importance of C-erb B2 expression in breast cancer. J Clin Oncol 1992;10:1049-1056.

28 Bianchi S, Paglierani M, Zampi G et al. Prognostic significance of C-erb B2 expression in node negative breast cancer. Br J Cancer 1993;67:625-629.

29 Press MF, Pike MC, Chazin VR et al. Her-2/neu expression in node-negative breast cancer: direct tissue quantification by computerized image analysis and association of overexpression with increased risk of recurrent disease. Cancer Res 1993;53:4960-4970.

30 Seshadri R, Firgaira FA, Horsfall DJ et al. Clinical significance of Her-2/neu oncogene amplification in primary breast cancer. J Clin Oncol 1993;11:1936-1942.

31 Descotes F, Pavy J-J, Adessi GL. Human breast cancer: correlation study between Her-2/neu amplification and prognostic factors in an unselected population. Anticancer Res 1993;13:119-124.

32 Giai M, Roagna R, Ponzone R et al. Prognostic and predictive relevance of C-erb B2 and ras expression in node-positive and negative breast cancer. Anticancer Res 1994;14:1441-1450.

33 Muss HB, Thor AD, Berry DA et al. Cerb-B2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266.

34 Têtu B, Brisson J. Prognostic significance of Her-2/neu oncogene expression in node-positive breast cancer. The influence of the pattern of immunostaining and adjuvant therapy. Cancer 1994;73:2359-2365.

35 Hartmann LC, Ingle JN, Wold LE et al. Prognostic value of CerbB2 overexpression in axillary lymph node-positive breast cancer. Results from a randomized adjuvant treatment protocol. Cancer 1994;74:2956-2963.

36 Jacquemier J, Penault-Llorca P, Viens P et al. Breast cancer response to adjuvant chemotherapy in correlation with erb B2 and p53 expression. Anticancer Res 1994;14:2773-2778.

37 Marks JR, Humphrey PA, Wu K at al. Overexpression of p53 and Her-2/neu proteins as prognostic markers in early stage breast cancer. Ann Surg 1994;219:332-341.

38 Rosen PP, Lesser ML, Arroyo CD et al. Immunohistochemical detection of Her-2/neu expression in patients with axillary lymph node-negative breast carcinoma. A study of epidemiologic risk factors, histologic features and prognosis. Cancer 1995;75:1320-1326.

39 Quénel N, Wafflart J, Bonichon F et al. The prognostic value of c-erbB2 in primary breast carcinomas: a study on 942 cases. Breast Cancer Res Treat 1995;35:283-291.

40 Sundblad AS, Pellicer EM, Ricci L. Carcinoembryonic expression in stages I and II breast cancer; its relationship with clinicopathologic factors. Hum Pathol 1996;27:297-300.

41 O’Malley FP, Saad Z, Kerkvliet N et al. The predictive power of semiquantitative immunohistochemical assessment of p53 and C-erbB2 in lymph node-negative breast cancer. Hum Pathol 1996;27:955-963.

42 Hieken TJ, Mehta RR, Shilkaitis A et al. Her-2/neu and p53 expression in breast cancer: valid prognostic markers when assessed by direct immunoassay, but not by immunochemistry. Proc Annu Meet Am Soc Clin Oncol 1996;15:113a.

43 Xing W-R, Gilchrist KW, Harris CP et al. FISH detection of HER-2/neu oncogene amplification in early onset breast cancer. Breast Cancer Res Treat 1996;39:203-212.

44 Dittadi R, Brazzale A, Pappagallo G et al. ErbB2 assay in breast cancer: possibly improved clinical information using a quantitative method. Anticancer Res 1997;17:1245-1247.

45 Fernandez-Acenero MJ, Farina Gonzalez J, Arangoncillo Ballesteros P. Immunohistochemical expression of p53 and c-erbB-2 in breast carcinoma: relation with epidemiologic factors, histologic features and prognosis. Gen Diagn Pathol 1997;142:289-296.

46 Eissa S, Khalifa A, el-Gharib A et al. Multivariate analysis of DNA ploidy, p53, c-erbB-2 proteins, EGFR, and steroid hormone receptors for short-term prognosis in breast cancer. Anticancer Res 1997;17:3091-3097.

47 Charpin C, Garcia S, Bouvier C et al. c-erbB-2 oncoprotein detected by automated quantitative immunocytochemistry in breast carcinomas correlates with patients’ overall and disease-free survival. Br J Cancer 1997;75:1667-1673.

48 Press MJ, Bernstein L, Thomas PA et al. Her-2/neu gene amplification characterized by fluorescence in situ hybridization: poor prognosis in node-negative breast carcinomas. J Clin Oncol 1997;15:2894-2904.

49 Ross JS, Muraca PJ, Jaffe D et al. Multivariate analysis of prognostic factors in lymph node negative breast cancer. Mod Pathol 1998;11:26a.

50 Depowski PL, Brien TP, Sheehan CE et al. Prognostic significance of p34cdc2 cyclin dependent kinase and MIB1 overexpression, and HER-2/neu gene amplification detected by fluorescence in-situ hybridization in breast cancer. Mod Pathol 1998;11:18A. 

Academia and advocacy conflict on economics of malaria eradication

[Please see the clarification at the end of this article.]

Don’t ask where the numbers came from, but beating malaria is a great way to make $4 trillion dollars.

In July, the Rollback Malaria initiative rolled out its “exceptional case” for investment in eliminating malaria, a plan promising a 40:1 return on investment (ROI), rising to 60:1 in sub-Saharan Africa.  Malaria elimination will purportedly bring a multi-trillion dollar windfall: $700 billion within a few short years (by 2020), growing to an impressive $4.1 trillion by 2030.

(Source: Rollback Malaria, Action and Investment to defeat Malaria 2016-2030)

However previous research, also conducted by advocates for malaria elimination, found insufficient basis for using financial benefits to justify the costs of fighting malaria.

Rollback Malaria (RBM) would not explain how it arrived at its inviting rates of return, declining to provide a spreadsheet. “The Excel sheet has data from all countries that was modelled up to give the global costs so it isn’t terribly helpful,” said Helen Prytherch of the University of Basel. Prytherch would not send the spreadsheet and suggested talking with the lead economist instead.

The process of peer review prevented bringing forth the details of the estimate. “We are working on your request,” said Prytherch, who continued:

“Several scientists from different institutions worked over a two-year period to establish the new malaria targets, cost them and then to establish and implement a cost benefit analysis. They are now preparing scientific papers to get the work into the public domain. They should be ready for submission by end September.”

However, in 2010, a consortium of pro-elimination researchers concluded that "policy makers should not view the generation of substantial short-term or medium-term cost-savings as a rationale for elimination until more robust evidence is available to suggest otherwise."

Investigators at the University California at San Francisco (UCSF) subsequently sought but did not find robust evidence of cost saving. Findings from ten case studies “do not change the conclusion” that cost should not be viewed as rationale for elimination, according UCSF’s Rima Shretta. Shretta said “malaria elimination should not be pursued merely on the grounds of cost-savings as these often fail to capture all the externalities garnered by disease elimination.”

Of RBM’s plan, Shretta says “it is an advocacy document rather than an academic analysis.” RBM’s macroeconomic analysis “was done using published data rather than an empirical analysis.” In the reinterpretation of existing results, RBM “used a full-income approach which produced larger benefits than some other approaches.” Several previous studies “have not been able to do this using short term projections,” according to Shretta.

Either malaria advocacy is going to depart from established academic research, or the academic consensus might be about to change. “I think if an academic analysis supports elimination – great,” said Shretta. Rather than argue against such an analysis, Shretta seems to have adapted to its inevitability: “the benefits are often underestimated so [elimination] should not ONLY be based on an economic argument but ALSO a social, development and moral perspective.”

On the present trajectory, science will soon give its blessing to malaria eradication being wildly profitable in addition to its other virtues.

Clarification

Posted August 26, 2015

Since publication of this article, Rima Shretta disclosed to me her membership in the task force for Action and Investment to defeat Malaria (AIM). AIM produced the estimate of $4.1 trillion of economic benefits to be gained from malaria elimination examined in my article. Shretta would not explain why she did not inform me of this critical fact. However, it is the responsibility of the journalist to establish the independence of an observer and I regret that I failed in this regard.

Shretta disputes multiple aspects of the article. Prior to disclosing her involvement in AIM, Shretta requested a correction to the article:

I believe my opinion has been mis-represented. I do not dispute the analysis in AIM. I also have not adapted to its inevitability. The statement below is incorrect:

"Rather than argue against such an analysis, Shretta seems to have adapted to its inevitability"

We strongly believe in the case for elimination and its returns - financial and otherwise. In my opinion often the analysis does not make the case strong enough because of the other factors that cannot be measured well.

I responded in part:

I asked Richard Feachem if there was new evidence on the matter of costs and benefits. He referred me to you. You stated there was no new evidence to change the conclusion of the 2010 [Lancet] paper.

So something has to give. There is a $4 trillion discrepancy. 

Shretta, now stating she was on the AIM task force, then requested multiple changes to the article, including re-working or deleting all quotations except one. She wrote in email: “I am fine with the statement that ‘malaria elimination should not be pursued merely on the grounds of cost-savings as these often fail to capture all the externalities garnered by disease elimination.’ “ However, “the rest of the quotes are out of context…”

Shretta did not reply to a question regarding her involvement with AIM. However, I withdraw the statement that Shretta adapted to the inevitability of the AIM document.

Post-publication, regarding the key question of reconciling the Lancet paper with AIM, Shretta wrote: “The Lancet paper warned on rationalizing elimination based on a financial argument alone.” Arguably, this is a considerable misinterpretation. The Lancet paper warned that short- and medium- term economic benefits cannot be part of an argument for elimination, whether alone or in conjunction with any other arguments.

Shretta’s penultimate communication states:

The AIM is based on country level data on malaria disease and a transmission model that forecasts the disease over time. This was then costed out. The analysis on the benefits is from published data. We at UCSF do not dispute the findings of AIM.

One less apocalypse: Questioning the spread of drug-resistant malaria

Drug resistance has twice started in Southeast Asia, both times leading to massive epidemics of untreatable malaria in Africa. Only the introduction of artemisinin combination therapy earlier this century beat back the most recent wave of drug resistance. Now artemisinin is buckling, leading to understandable worry about yet another resistance apocalypse. But current scientific evidence contradicts the narrative of doom voiced by journalists (including me) and much of the malaria research community.

Artemisinin resistance has barely spread but instead popped up on its own, evolving independently in areas scattered across Southeast Asia. The effort to “contain” resistance by wiping out malaria in the region will not prevent independent emergence in Sub-Saharan Africa where home-grown resistance could develop undetected by today’s weak surveillance system.

Artemisinin resistant malaria results from changes to a complicated genetic network that will be difficult to infiltrate into other parasite populations without it coming apart. However, elimination efforts Southeast Asia both strengthen resistance and streamline its genetics for easier transmission abroad, fomenting the very apocalypse it supposedly seeks to avoid.

Not spreading even in Southeast Asia

Drug resistant malaria is scarcely spreading at all in Southeast Asia, even within national borders. The most comprehensive survey found only three instances of spread out of 112 samples from across the region. Parasites thought to have originated in Cambodia were found in people tested near the border with Vietnam. “All other mutations appear to have arisen independently,” scientists concluded. Other researchers concur that resistance “is primarily due to the proliferation of newly emerging mutations…” Rather than spreading, most instances of resistance “appear to be localized to a relatively small geographical area…”

A paper on resistance in Myanmar includes the word “spread” in its title but adduces little evidence and no claims for it. Scientists found “strong evidence” of resistance in Myanmar “including regions close to the Indian border in the northwest,” a worry because past drug resistance is thought to have spread first to India before leaping to Africa. However, resistance is not spreading, according to the authors, rather it “extends” across Myanmar. Seven individual mutations appear “to have arisen independently” more than once, pointing not to spread but de novo emergence.

Resistance has not spread to Myanmar: “Contrary to the widely assumed scenario,” concluded another research group, “we found no evidence of westward spread of artemisinin resistance from Cambodia to Myanmar.” So far in Laos and Bangladesh, mutations associated with resistance are “absent or found at much lower frequency,” additional evidence against regional spread.

Sizable population movements within and between countries seemingly ought to create an equivalent dispersion of resistance. Why that hasn’t happened is “a very good question,” said Philippe Guyant, co-author of a paper on malaria and migrant workers in Cambodia. “I don’t think there is a definitive answer to it given the current state of knowledge.”

But the domino theory of spreading resistance, although widely-discussed and deeply worrisome, is not supported by current scientific evidence which shows that drug-resistant malaria is scarcely spreading even within Southeast Asia.

Genetic complexity militates against spread

Conclusive evidence of artemisinin resistance emerged in 2008, but the complexity of the underlying genetics frustrated efforts to find a molecular marker until 2014 when mutations in a gene called Kelch 13 (K13) were finally implicated.

K13 mutations appear necessary but not sufficient for resistance; supporting mutations appear to be needed. Scientists inserted resistance-associated K13 mutations into parasites susceptible to artemisinin. Modified parasites originally from Cambodia showed a greater increase in resistance than other genetically altered lines, “suggesting a role for additional parasite factors in augmenting K13-mediated resistance…” Four other genes have been connected with resistance in Southeast Asia. In two African samples, scientists found the supporting mutations “were rare or absent… suggesting that they are the product of evolutionary selection within Southeast Asia.” In Southeast Asia, the K13 mutations appeared only after the supporting cast was in place.

Right now, artemisinin resistance in Southeast Asia is tightly bound to an interconnected set of genetic changes particular to its evolutionary history, a history that differs greatly from much of Africa—although not all.

For artemisinin resistance to spread to Africa it will have to overrun incumbent populations. However, the delicate architecture of resistance—multiple mutations riding several different chromosomes—is likely to be pulled apart by the sexual recombination of malaria parasites. Years ago, malaria dragged down the combination drug sulfadoxine-pyrimethamine by incrementally accumulating changes. By contrast, a mutation to K13 seems to need simultaneous changes elsewhere in the genome to balance fitness costs, compete with other parasites or both.

According to Olivo Miotto, co-author of a paper on the genetic architecture of artemisinin-resistance:

“The fact that the main Kelch 13 mutations emerge only on a certain genetic background suggests that there is something special about those parasites. Perhaps it is this ‘something special’ (associated with the genetic background we have identified) that needs to spread in Africa before Kelch 13… I’m pretty sure that Kelch 13 mutations alone will not be enough.”

Another obstacle to the spread of resistance to Africa is far lower drug pressure there. Malaria in Southeast Asia is less intense, so people generally do not acquire immunity and fall ill when infected. The sick seek and receive treatment at very high rates, piling on drug pressure. By contrast, “Right now, only a portion of African parasites get exposure to the drug,” observed Miotto. In much of Sub-Saharan Africa, more intense malaria means greater natural immunity, leading to a greater number of infections that cause no sickness. The unsick seek no treatment. If parasites aren’t exposed to the drug, resistance to artemisinin confers no fitness advantage and will be swept from the genome.

Unknown unknowns

However, much remains unknown and possible parallels with the past are cause for concern. Although artemisinin resistance appears to rely on more than K13 mutations, according to Miotto, chloroquine resistance, early in its development, also might have needed more than one mutation. For chloroquine, “the key marker was identified, but the story may still be incomplete—don’t confuse that with it being simple… We could only study the aftermath,” which pointed to a mutation in one gene. According to Miotto, malaria could yet produce a single K13 mutation that prevails against artemisinin.

Chris Plowe at the University Maryland concurred: “What is happening now with artemisinin resistance may not be all that different from what happened with chloroquine, sulfadoxine and pyrimethamine resistance in the past. We are just witnessing it in real time with a lot more data.”

Continuing research might find more evidence for spread. According to Shannon Takala-Harrison at the University of Maryland, with increasing numbers of samples, “we are seeing additional evidence for spread as well as independent emergence of mutations.” She looked forward to discussing “more concrete results and conclusions as they become available.”

At this particular time, however, there is astonishingly little evidence of spread and sizable genetic and environmental obstacles working against it.

Policy discomfited by evidence

The current policy of elimination fits somewhat awkwardly with current evidence. If artemisinin resistance mostly emerges independently, increased surveillance in Africa rather than containment in Southeast Asia might be more sensible. But according to Patrick Kachur, head of the malaria branch at the Centers for Disease Control (CDC), eliminating all malaria in Southeast Asia makes good sense: “I think the threat of artemisinin-resistance spreading to Africa is a compelling reason why global malaria advocates should be interested in eliminating malaria in Southeast Asia.” Also, countries in the region have their own, additional reasons for wanting to be completely free of malaria, according to Kachur.

Elimination has been “a moderately effective advocacy message,” Kachur said. It is one the malaria research community seems loathe to change even though science does not clearly support it. Neither Kachur nor Plowe responded to emails asking them to contradict the hypothesis that drug-resistant malaria is not actually spreading. Emails to the Gates Foundation also received no answer.

The narrative of doom obscures an all-too rare bright spot in malaria and global health: the pipeline for new antimalarial drugs is incredibly robust. Although grim headlines say, for example, “No 'plan C' drugs available,” multiple new candidate drugs and entire new drug classes have been discovered largely under the umbrella of the Medicines for Malaria Venture, a public-private partnership started by the Gates Foundation.

However, Bill Gates adds his voice to the apocalypse chorus. In a YouTube video, Gates described the possible spread of resistance as “the biggest disaster for control ever.”

Next Gates says: “We’re trying to figure out if we can do local eradications.” But if resistance, rather than eradication, were the primary concern, elimination is no longer automatically the right strategy.

“Drug resistance is driven by drugs,” as Olivo Miotto put it. Elimination maximizes drug resistance, making it stronger and more heritable as drug pressure reshapes and streamlines the initially complex genetics of resistance. This is “the core question” for Miotto. “Drug resistance doesn’t come from heaven; we create it, we encourage it.” He called for better models “to predict the outcome of intervention as we move forward,” saying “responsible approaches to deploying drugs are key.”

Nonetheless, CDC’s Kachur said “enthusiasm is high among global and subregional malaria subject matter experts” for elimination.  Chris Plowe argues that “What the independent emergences tell us is that containment is not likely to work, so by eliminating we can at least try to prevent the most fit, viable and dangerous forms from spreading.” However, elimination propels greater fitness, viability and more dangerous forms that are more likely to spread.  Per the title of a 2009 paper about eliminating artemisinin-resistant malaria in Cambodia, “The last man standing is the most resistant.”

Mathematical models show elimination is unlikely to work. Gates Foundation-funded researchers found that extinguishing malaria was not possible in many places, including in Southeast Asia: “Prospects for elimination in Myanmar and southern Thailand do not appear to be favorable.” Myanmar, recently announced the goal of eliminating malaria.

In another study, “An optimal control strategy to reduce the spread of malaria resistance,” even models using both mass drug administration and insecticide measures fail to completely get rid of drug-resistant malaria. “We think from our models that it is true it is not possible to eliminate drug resistant malaria just using mass drug administration and insecticide,” confirmed co-author Fatmawati Armawi of the Universitas Airlangga.

With elimination exacerbating resistance, evidence-driven policy would seem to suggest reducing drug pressure in Southeast Asia and intensifying surveillance in Africa. The edges of the malaria belt in Africa have low transmission like Southeast Asia. In addition, countries that have advanced toward malaria elimination also have low transmission coupled at times with high drug pressure. According to Miotto, “These should probably be our ‘sentinels’ ” for artemisinin resistance in Africa.

At present, however, malaria science and malaria advocacy appear to have separated.

Malaria elimination efforts jeopardizing early pregnancies in Southeast Asia

Bill & Melinda Gates in Pailin, Cambodia (Photo/video still: Gates Foundation)

Large-scale drug administration campaigns are putting early pregnancies at risk in Southeast Asia where efforts are under way to eliminate malaria. World Health Organization (WHO) treatment guidelines state that frontline antimalarial drugs based on artemisinin should not be given to women in the first trimester of pregnancy. Animal studies have found artemisinin caused early termination of pregnancies and birth defects.

But few programs test for pregnancy, according to the US Centers for Disease Control (CDC). Even a malaria treatment project funded and visited by Bill & Melinda Gates in Pailin, Cambodia seems not to be screening for pregnancy and departing from WHO guidelines.

Eliminating drug resistance & gearing up for global eradication

In Southeast Asia, the countries surrounding the Mekong River are seeking to completely eliminate malaria. The driving force comes from concern that drug-resistant malaria might spread from Asia to Africa, which has happened twice in the past at enormous human cost. Now artemisinin is under threat. In addition, elimination efforts in the Mekong region can provide valuable experience for the much greater ambition of global malaria eradication. As Bill Gates put it, “We’re trying to figure out, can we do local eradications?”

Malaria elimination leans heavily on large-scale administration of the frontline antimalarial drugs, artemisinin combination therapy (ACTs). Some campaigns test for infection, the “screen & treat” approach. Other campaigns simply treat everyone regardless of infection status in mass drug administrations (MDAs).

“It’s not possible to generalize,” how drug campaigns handle pregnancy, according to Patrick Kachur, malaria branch chief at the CDC. There are many campaigns and multiple institutions behind them, sometimes in partnerships. According to Kachur:

"In some of the MDA trials or pilot programs currently pregnant women were excluded by design.  In others they were not (or that detail has not been reported).  In most of the test and treat approaches pregnant women were usually included (occasionally receiving a different treatment regimen than children and non-pregnant adults if they tested positive)."

As a result, women who are or might be in the first trimester of pregnancy are being given artemisinin in some campaigns. Some pregnant women treated for malaria might not even be infected with the disease. WHO guidelines call for quinine and clindamycin in the first trimester of pregnancy--when the mother actually has malaria.

Artemisinin appears to be safe for mothers in all stages of pregnancy. However, in animals, artemisinin is embryotoxic and causes birth defects. (See review here.) The animal exposures to artemisinin were not extreme but adjusted to be near the equivalent, WHO-recommended therapeutic dose for humans. Even so, animal models can be misleading. The shorter development period in rats might be far more sensitive to artemisinin exposure than the more prolonged development process in humans and that “could have a protective effect for human fetuses,” as one researcher noted. Artemisinin might be safe—or not.

Assessing risk: prioritize obstetrics or malaria control?

In 2007, researchers wrote that larger, “methodologically rigorous” studies of artemisinin and pregnancy were “urgently required.” The authors worried that “early pregnancy loss will be difficult to detect, especially in communities where artemisinins are likely to be used most frequently.”

But more recently, concerns have partly subsided, perhaps more among malaria specialists than obstetricians. “My concern has gone down on this issue,” said Brian Greenwood, of the London School for Hygiene and Tropical Medicine and co-author of the 2007 paper calling for examination of artemisinin safety. More recently, Greenwood said: “There is now extensive clinical experience that ACTs are safe in the second and third trimesters but, not surprisingly, less data on exposure in the first trimester.” 

There has been no larger, methodologically rigorous safety study; it might not be possible to perform ethically. Instead, “the numbers of documented cases of exposures in the first trimester is still fairly limited,” said Greenwood, “in the hundreds, so a rare event could not be excluded and it would be difficult, or probably impossible, to detect fetal resorption.” Fetal resorption is defined as “The disintegration and assimilation of the dead fetus in the uterus at any stage after the completion of organogenesis which, in humans, is after the 9th week of gestation.” 

The Gates Foundation, asked whether artemisinin posed a health risk in early pregnancy, demurred. Foundation spokesperson Bryan Callahan instead suggested seeking comment from WHO “on whether they are planning to revise their normative guidance.” Callahan expected that WHO “would take available scientific research into account in reviewing their guidance, including a growing body of observational research on pregnant women.” Meetings in coming months could see the WHO guidelines revised.

However, the safety of artemisinin in early pregnancy is not established by evidence that would lead to regulatory approval in the developed world. Physicians in the United States would not administer artemisinin to a pregnant woman in the first trimester, particularly in the absence of a malaria infection, as is happening in countries like Cambodia and other nations in the Mekong River region.

Wealthy countries don’t have malaria and so can prioritize pregnancy. Still, a public health policy that increases pregnancy risks to mothers living with less money and more disease makes for a problematic ethical situation at best.

'Programs should screen for pregnancy'

“I think programs that use MDA should provide pregnancy testing like we do in Wellcome Trust units,” said Rose McGready from the Shoklo Malaria Research Unit in Mae Sot, Thailand. According to McGready, proving safety in first trimester drugs or vaccines “is extremely difficult and more so in countries where health systems are not working well.”

Even regarding currently approved drugs, McGready asked: “how much data do we have for them? Many are assumed to be safe [like] quinine; but only proper comparative studies will provide a definitive answer.”

Melinda Gates has been campaigning for “Putting women and girls at the center of development,” as she wrote last year in Science. According to Gates, the foundation focused in its earliest days on research. Its second phase included an emphasis on delivery. For the foundation's third incarnation, “what I’m making sure we add on now is the women and girls lens,” she recently said.

But that lens seems to have been absent when Melinda and Bill Gates visited a screen and treat program in Pailin, Cambodia earlier this year.

Blogged Bill Gates:

“we walked to a local school where the screening is taking place. That morning, about 120 people had come to get their blood drawn and tested for the malaria parasite. They also answered a few questions designed to find out whether they might have been exposed to the parasite (e.g., ‘Do you work in the forest?’).” 

Gates did not mention questions about pregnancy or pregnancy tests.

Dance of the blameless

Asked whether the Pailin program included pregnancy screening, foundation spokesperson Bryan Callahan replied: “We recommend that you direct any detailed questions to MORU,” Mahidol Oxford Tropical Medicine Research Unit. MORU was the foundation partner responsible for the project and orchestrated the Gates’ visit to Pailin. According to Callahan, “Like all foundation grantees, MORU was required to secure country-level IRB approval for its malaria treatment protocols, and these protocols include a pregnancy screening component.”

Callahan would not confirm that other Gates grantees were screening for pregnancy, although he acknowledged that he had "received the feedback that I had requested from partners" as part of what he termed "due diligence" in answering the "chemotherapy for pregnant women question."

Callahan would not provide a list of the Gates Foundation partners. “We list all of our Malaria program grantees on our website, and you are free to contact them,” said Callahan. A search for “malaria” on the foundation’s grant website returns 1,000 matches. 

Asked whether MORU specifically was testing for pregnancy rather than just required to, Callahan answered: “The partner is MORU, so you have an answer to your question.” The answer, however, was not “yes.” Pressed further, Callahan said: “As I have stated several times, foundation grantees are required to use protocols approved by local IRBs. You need to consult directly with MORU on your question.”

Asked for the most appropriate contact at MORU, Callahan supplied a link to the MORU contact page.

Buck passed

According to MORU’s Lorenz Von Seidlein, “We are coordinating several studies which include mass drug administrations and are funded by the BMGF,” the Bill & Melinda Gates Foundation. Regarding the scope of the effort, Von Seidlin wrote: “drugs have been administered in Vietnam and… in [the] Thai-Myanmar border areas [while] drug administrations are planned in Pailin/Battambang Cambodia in the coming weeks and in Laos at the beginning of next year.”

To describe the project, Von Seidlin pointed to a paper entitled “Fighting fire with fire.” It likened targeted malaria elimination to the tactic of “back burning” in battling forest fires. According to the paper, “all community members whether infected or not are offered antimalarial treatment.” The three-day treatment is given a minimum of three times, one month apart, creating multiple possible exposures of first trimester pregnancies. (It’s not clear that such a regimen has been tested in animal models. Some animal studies found pregnancy harms from artemisinin increased with dose size.)

The paper does not mention pregnancy screening. Asked in email, “Are the mass drug administrations screening for early pregnancy?” as the Gates Foundations says is required of its partners, Von Seidlen did not reply.

Echo?

The nonprofit FHI360 is administering a malaria grant from the Global Fund, also focused on Pailin. Asked if pregnancy screening was part of the program's protocol, an FHI360 spokesperson "reached out to our experts" but never replied to the question. FHI360 touts its namesake "360° perspective" and lists "gender" as a practice area.

WHO's Walter Kazadi coordinates the Emergency Response to Artemisinin Resistance (ERAR) in the Greater Mekong Subregion. Kazadi did not reply to email asking about anti-malarial administration and pregnancy screening.

Eradication: an experiment

Whether the drug-based strategy will eliminate malaria is not known. According to Von Seidlen’s paper, “It is not clear what coverage is required to interrupt transmission, a question mathematical modelers may be able to answer.” However, Gates-funded modelers have already said mass drug administration alone will not eliminate malaria in Southeast Asia.

Bednets and insecticide spraying will be hard pressed to close the gap as substantial numbers of people at risk for malaria live largely outdoors. Many do not wish to be offered malaria treatment or even to be found by government or non-government organizations. To gain cooperation in relatively docile Pailin, Bill Gates said those participating “were paid a day’s wages, the equivalent of about $2.50, and got a free lunch.”

About 1,700 people were processed, but the program would need to be scaled up to reach 4 million people in Cambodia, according to Gates. “We have to clear the parasites of all the humans in an area,” Gates said, making no exclusion for pregnancy. “Eradication is an ambitious goal,” concluded Gates. “It is a goal to which we remain 100% committed.”

'Radical cure' and pregnancy

Pregnancy might pose some difficulties for eradication. The Gates Foundation’s strategy calls for a “complete cure,” a new drug able to clear malaria infections in one dose, unlike today’s three-day regimen. However, the more radical the cure, the greater the potential impact on pregnancies.

Fortunately, one leading candidate, OZ439, looks far better than artemisinin: “OZ439 is 100 times safer,” according to Tim Wells, Chief Scientific Officer at the Medicines for Malaria Venture (MMV). Wells did not point to a paper or adduce evidence for his statement.

Another highly promising drug, KAE609, presents more of a mystery—even to Wells. Although KAE609 originated from a partnership of MMV and Novartis, the drug company re-possessed its intellectual property after discovering the considerable promise and commercial prospects of KAE609. The rest of the world and even Wells are now on the outside looking in. Novartis apparently has safety data but “has not talked about them externally,” according to Wells.

In early studies, KAE609 was given in multiple smaller doses: three times, 30 milligrams per dose, “which gives a certain plasma exposure,” said Wells. More recently, aiming for radical cure, a single dose of 75 milligrams has been tested. “If they have to go with the higher number," 2.5 times higher, "the safety margin is of course a little bit lower,” observed Wells.

Two other drug candidates are in development, providing a quite remarkable and impressive range of options. “The key will be that we can’t design molecules safe for pregnancy," said Wells, "but we can at least pick the most likely candidates, now that we have a little bit of choice.” 

The choice will be important. More mass drug administrations are likely. According to Bernard Nahlen, "the countries which have eliminated up to this point have not done so without MDA." Nahlen is the Deputy Coordinator of the President's Malaria Initiative. Also, malaria diagnostics aren’t sensitive enough to find low level infections. To clear every infection, including those that are undetectable, eradication would mean “treating” even the uninfected and the possibly pregnant. According to Wells, “for MDA where the subjects don’t have the disease, we need to be looking at vaccine levels of safety – say one serious adverse event in 20,000 cases.” 

However, given current practices which elide or ignore pregnancy concerns in Southeast Asia, global malaria eradication might expose much of a generation in Sub-Saharan Africa to antimalarials, whether artemisinin or new drugs in the pipeline, whose effects on pregnancy and development are not fully understood.


Article history:

[7/22/2015 2:46 PM] Quotation from Bernard Nahlen added

Vaccine-associated polio: ignored, set to rise?

Bill Gates administering oral polio vaccine in Chad (Photo: Gates Foundation)

Thanks to oral polio vaccine, the world has nearly extirpated a crippling disease from the planet. In rare instances, however, the same vaccine can cause polio. With progress in eradication, vaccine-associated cases of paralysis began to surpass cases caused by the disease in 2012. A switch in oral vaccines next year might increase vaccine-induced paralytic polio. An inactivated version of the vaccine is available that cannot cause polio and can prevent the polio that infrequently results from the oral vaccine. However, the inactivated formulation is only now being rolled out and not in a way that will stop the oral vaccine from sometimes causing paralytic polio.

Schedule and budget appear to be driving polio policy, not minimizing cases of paralysis from all sources, including the oral vaccines. 

The last case of paralytic polio in the world might be caused by the live oral vaccine.

A devil's bargain comes into view

The risk of “vaccine-associated paralytic polio” (VAPP) is very low: 3-4 cases per million births, according researchers at the World Health Organization (WHO) and US Centers for Disease Control (CDC). But with so many children immunized with it, the oral vaccine caused an estimated 399 cases of paralysis in 2012 compared to just 223 caused by polio itself. This disparity will only worsen as eradication proceeds. Schedule slips will mean not only more money (perhaps $1 billion a year) but also hundreds more polio cases caused by the oral vaccine.

Mutations in the live vaccine virus can cause not only VAPP but lead to infection of others, just like the wild virus. Madagascar, for example, has recently seen multiple cases of polio from circulating vaccine-derived viruses. Pakistan and Nigeria have also been battling transmission of polio virus that came from the oral vaccine.

As a first step toward complete cessation of live vaccine use, current plans call for the trivalent oral polio vaccine (tOPV) to be withdrawn worldwide next April, replaced by a bivalent vaccine which immunizes against only types 1 and 3 of the poliovirus. (Type 2 appears to be long gone, last seen in India in 1999.) However, the bivalent vaccine could increase VAPP cases.

More VAPP or less?

The data are scant but concerning. Experience in Hungary “suggest a higher rate of VAPP associated with the use of bivalent OPV compared to tOPV," according to researchers at the CDC and WHO, 20 times higher. However, the data are limited, seemingly to one year, 1961.

VAPP risk varies widely depending on context. According to the CDC, the “best data on VAPP” for the monovalent oral polio vaccines “comes from Hungary, where these strains have been used the longest.” In addition, Hungary featured excellent detection and investigation, requiring that every suspected case of poliomyelitis be admitted to a central hospital for clinical and laboratory evaluation. However, these practices only came into full effect in 1966, five years after the 1961 administration of the bivalent vaccine that generated so many cases of VAPP.

Not only Hungary, but Belarus and especially Romania reported unusually high rates of VAPP, as many as one case per 183,000 doses. However, research published in the high-profile New England Journal of Medicine put these concerns to rest, attributing VAPP in Romania largely to “provocation paralysis,” or multiple intramuscular injections administered shortly after oral polio vaccination. However, some of the same researchers subsequently found that in the United States, intramuscular injections did not cause VAPP, results published to less notice in the Pediatric Infectious Diseases Journal

The dismissal of higher VAPP rates in parts of Eastern Europe, however, still stands. According  to WHO and the CDC: “There is no evidence that the high risk of VAPP observed in these studies is representative of the risk of VAPP in the majority of OPV-using countries globally.”

Most cases of VAPP are caused by the type 3 vaccine virus. But the trivalent vaccine causes less VAPP than the type 3 monovalent vaccine. Analysis of US data from the 1960s and 70s found that the trivalent vaccine halved the risk for VAPP, perhaps suggesting that the trivalent formulation has a taming effect on type 3 VAPP.

The type 2 vaccine virus is actually alpha dog, outcompeting both the type 1 and type 3 viruses of the trivalent vaccine when it comes to infecting (usually benignly) the body. Global health authorities expect that dropping type 2 from the vaccine will greatly reduce VAPP: “removal of type 2 serotype from OPV provided globally in routine immunization and campaigns could decrease the overall risk of VAPP by at least 25%–30%.”

However, just subtracting out the percentage of VAPP cases attributable to the type 2 component of the vaccine might be overly simplistic. The presence or absence of type 2 clearly impacts the effects of type 3 in the body. Leave out type 2 and better protection for type 3 results, for example. The only available evidence—the limited data from Hungary—points to much higher VAPP from bivalent than trivalent vaccine.

Anti-vaxxers' delight

Roland Sutter, a scientist at the World Health Organization and co-author of numerous of papers on VAPP, dismissed out of hand that bivalent vaccine might increase VAPP, saying: “I don’t believe anything that hasn’t been proven.” Sutter pointed out that four billion doses of the bivalent vaccine have been administered since 2009 and "no safety signal has been detected anywhere in the world.” He asked: “Wouldn't you see something?"

However, WHO might see no safety signal because WHO doesn’t track VAPP. “The countries are keeping track,” according to Sutter. The bivalent vaccine “does cause VAPP as well," Sutter said, but identifying VAPP cases is technically demanding. “It’s not so easy to go through the algorithms," he explained. And countries, perhaps like WHO, may have little incentive to track and report how many children and adults are being paralyzed by a public health program. Asked if WHO had a spreadsheet aggregating country-level VAPP data, Sutter replied: “Not at all. No.”

VAPP: preventable, like polio

In theory, VAPP could be avoided entirely by using the inactivated polio vaccine (IPV). The United States dropped the live oral vaccine in 2000 “to eliminate the risk of vaccine-associated paralytic poliomyelitis (VAPP),” according to the CDC. Most wealthy countries immunize with IPV. But IPV poses a number of problems for eradication.

IPV must be injected, whereas a deluge of oral vaccine drops can be unleashed by armies of untrained vaccinators. The high levels of vaccine coverage needed, over 90%, would be much, much harder to attain if polio eradication relied on national routine immunization programs which can handle injections. The eradication effort opted for oral vaccines and also for a separate, polio-only vaccination infrastructure that actually drew resources away from routine immunization programs.

IPV by itself also likely would not suffice to eradicate polio. The live and inactivated vaccines confer different kinds of immunity. IPV only protects against paralysis from polio, not infection. In 2013, Israel found widespread polio transmission in sewage samples. Because of the country’s high IPV coverage and a little luck, no cases of polio resulted. But polio still circulated. Israel resumed immunizing with OPV while continuing IPV. Because OPV prevents both disease and infection, transmission in Israel soon stopped, demonstrating not only sharp work by scientists and public health officials, but also that eradication with IPV alone may be impossible. On the other hand, the strongest individual and population immunity to polio results from vaccinating with both IPV and OPV.

Another obstacle to universal adoption of IPV has been cost. Until recently, IPV cost about $2 per dose versus $0.10 - 0.15 for the oral vaccine. However, in 2000 when the US switched to IPV, a generous gift from the Bill & Melinda Gates Foundation led to the founding of Gavi. Gavi sought, among other aims, to slash the time it took for a vaccine to get from the rich world to the poor. Thanks to Gavi, relatively expensive vaccines for hepatitis and rotavirus became more quickly available in the developing world—but not IPV. More recently, Gavi began rolling out its most expensive vaccine yet, for Human Papillomavirus (HPV), which can cost more than $100 in developed countries.

The cost of IPV rather than its safety benefits continue to be at the forefront in policymaking decisions. A recent paper from Gates Foundation and CDC researchers stated: “In the global polio eradication end game, the cost of IPV will need to be balanced with effectiveness.”

Gavi’s support for IPV only began in 2013 with the publication of the polio endgame strategy, according to Gavi spokesperson Rob Kelly. Vaccine safety was not the main driver. According to Kelly, "the primary purpose of an IPV dose in Gavi countries is to maintain immunity against type 2 poliovirus," after withdrawal of the trivalent vaccine.

Vaccine schedule and VAPP: out of order

Gavi’s recent support for IPV will have little or no impact on VAPP because the oral vaccine will be administered first. To prevent VAPP, IPV must come before OPV. Brazil moved away from an OPV-only schedule, putting two doses of IPV first with the goal of “preventing rare cases of vaccine-associated paralytic polio” and “ensuring equitable access to IPV,” i.e. not inflicting VAPP on the poor.

However, WHO recommends only a single dose of IPV after the oral vaccine. According to WHO, children will then be older and maternal antibodies less likely to interfere with developing immunity in response to the vaccine. However, the CDC found that coverage with one dose of IPV “is expected to be lowest” when given on WHO’s recommended schedule and highest if given the first time a child is immunized. About 12 million children won’t get IPV if WHO’s plan is followed, according to the CDC.

But nations supported by Gavi will be following WHO guidelines, according to Gavi’s Rob Kelly: “countries have overwhelmingly decided to introduce the IPV dose at 14 weeks of age,” after the oral vaccine. The Gates Foundation supports WHO’s guidelines: “There are valid scientific and economic reasons why most Gavi countries still give OPV before the dose of IPV (generally at 14 weeks),” said foundation spokesperson, Rachel Lonsdale.

The foundation has criticized lags in rolling out vaccines in low-income countries but sees the handling of polio vaccines as similar “to what happened in the US,” according to Lonsdale. “When the risk of OPV is outweighed by the benefit the global program is moving to IPV.” Lonsdale emphasized: “We would not be where we are today and so close to eradication without OPV.”

Vaccine research arrives late

However, we would be much closer to eradication if there were a genetically stable oral vaccine. Such a vaccine would cause no VAPP and no circulating vaccine-derived virus. There would be no need to rollout the needle-based IPV.

Gates Foundation research into a vaccine with the safety of IPV and the infection-prevention of OPV began in 2011, according Lonsdale. She dates the foundation’s involvement with eradication to 2007 and a $100 million grant to Rotary International. The foundation became the largest financial backer of polio eradication in 2008. Scientists are also working on a genetically stable version of the oral vaccine but only more recently.

By contrast, in the early 2000s, the Gates Foundation pursued thermostable versions of many existing vaccines that required storage at low temperatures. The effort largely came to naught because breaking free of the vaccine cold chain required a thermostable version of every vaccine, with little or no benefit from converting just a few.

"Did you help that kid?"

The foundation's Lonsdale asserted that the global polio program "has always been concerned about VAPP." However, action on that decades-long concern has only come recently: "Due to the progress against WPV [wild poliovirus], VAPP is one of the major drivers in the 2013-2018 Endgame Plan to stop all OPV use by 2019," according to Lonsdale. Although VAPP is a driving concern, use of the live vaccine for a year after eradication of the wild virus means the last case of polio paralysis is likely to be caused by the oral vaccine.

The Gates Foundation hints that others have responsibility for choosing the two-edged sword of a polio vaccine that can cause polio: “for a more historical look at the history of polio vaccine policy, best to contact CDC or WHO,” Lonsdale suggested. 

Years ago, Bill and Melinda Gates showed their children a documentary about polio. The kids asked about a crippled boy in the film: "Did you help that kid? Do you know the name of that kid? Well, why not?" Melinda answered "We don't know that boy, but we're trying to help lots of kids like him." Bill reportedly added: "I'm in wholesale. I'm not in retail!"

VAPP is retail.

Complete detection: vaporware comes to malaria diagnostics

Rapid diagnostic test: 1,000 times too insensitive to detect all malaria infections (Photo: Wikimedia)

The second article in a series examining the pillars of the Gates Foundation’s malaria eradication strategy: Complete Detection, Complete Cure and Complete Prevention covered here.

Today’s rapid diagnostic tests (RDTs) are cheap, fast and easy to use: apply a pinprick of blood. Wait 15 minutes and read the result off visually, like a pregnancy test. RDTs distinguish malaria from other fevers and illnesses, leading to more appropriate treatment, improving both individual and public health. However, RDTs don’t detect malaria in people who aren’t sick but who still might have low level infections. Eradicating, rather than treating malaria, means finding every infection. But today “complete detection” is not practically feasible and might not be possible in theory. Because of these difficulties, detection might be set aside in favor of serial mass drug administration campaigns.

Next generation RDTs will hopefully be ten times more sensitive than current technology. Researchers also want the new RDTs to test for a second biological marker that signals the presence of malaria. Current RDTs check for a malaria protein called HRP2, but it is not expressed by all strains of the parasite. Screening and treating based only on HRP2 would select malaria parasites that are “resistant” to the diagnostic. But so far there is no consensus candidate for a second marker. Also, while the goal is a factor of 10 improvement, current plans call only for testing and confirming a 5-fold improvement.

A much larger problem, however, is that RDTs actually need to be 1,000 times more sensitive than they are today. Otherwise, the best alternative is to skip screening and instead treat everyone in mass drug administration (MDA) campaigns. A paper from Gates Foundation-funded researchers at Intellectual Ventures recently found that “Only diagnostics capable of detecting parasites below 0.1 parasites/microliter result in prevalence reduction on par with an MDA campaign.” Short by a factor of 1,000, current RDTs “are nowhere near sensitive enough and new technologies are necessary if MSATs [Mass Screening and Treatments] are to become the campaign of choice in the future.” Complete detection has a long way to go.

Paucity of alternatives

Such needle-in-the-haystack sensitivity can be had from a laboratory-based, molecular technique called PCR (polymerase chain reaction). PCR can even find a single parasite in a blood sample, a sensitivity of about 0.1 parasites per microliter of blood. But it currently requires laboratory conditions, expensive equipment, and trained technicians. RDTs cost around 50 cents. PCR equipment can cost $5,000 with individual tests running from $1.50 to $20 depending on the technology. Molecular diagnosis, in other words, is very expensive.

Also, a much larger volume of blood is needed, requiring a blood draw (and more highly trained staff) instead of a simple needle stick. PCR takes more time, about an hour or more. People might wander off before test results are in, especially in mass screenings when many or even most people won’t be ill. If blood samples are transported from the field to a centralized testing facility, they will need to be kept at 39 degrees Fahrenheit, according to current CDC guidelines, no mean feat in the high-temperature malaria belt. Considerable efforts are being made to make PCR more field-friendly but PCR cannot substitute for RDTs.

Few new technologies present themselves as alternatives. The Financial Times, in its annual World Malaria Day special section, dedicated an article to advanced diagnostic technology such as a tricorder-like device announced by Nanobiosym. The technology, according to the company, “allows you to diagnose any disease with a genetic fingerprint,” a compact, nanotech alternative to PCR. Company founder Anita Goel said Nanobiosym had not yet developed an “app” specific to malaria because market demand was uncertain. Goel said she had not spoken to the Gates Foundation. Asked to document a proof of concept for the technology, a Nanobiosym spokesperson said that information is “for the moment, highly confidential and proprietary and the company is only able to share under NDA.” Grand Challenges Canada supported a Nanobiosym trial to test for HIV in Rwanda but “there’s really no news to report,” said a Grand Challenges spokesperson. “There aren’t any results being published so far. They are still working on that.”

Two other groups mentioned by the Financial Times are working on a proven approach: magnetic detection of iron crystals called hemozoin. But while extremely clever, the method misses a large number of infections unless the blood sample is drawn at the right time. The approach was ruled out by Gates-funded researchers a year ago. Scientists from the University of Washington and Intellectual Ventures diplomatically concluded that they were “pessimistic about the diagnostic value of hemozoin-based methods at this time as a tool for malaria case management.”

Hemozoin detection (Photo: Intellectual Ventures)

In email, co-author Michael Hegg explained: “Many people have been (and continue to be) fooled by the ease with which hemozoin can be detected…” But it can be absent (or missed by current methods) even when malaria is present during the first part of the parasite’s lifecycle in humans. Consequently, testing people with moderate malaria infections for hemozoin will “miss more than 1 out of 10,” according to Hegg. The problem “only gets worse the fewer parasites there are to detect.” And eradication, as it proceeds, will result in fewer large infections and more smaller ones.

Indeed, “the last malaria reservoirs may the hardest to detect,” according to the motto of the DIAMETER project (Diagnostics for malaria elimination toward eradication). DIAMETER is tasked with finding next generation screening technology. It is managed by PATH and funded by the Gates Foundation. DIAMETER is a bit constricted, “not a very rich pipeline compared to vaccines,” according to Paul LaBarre who heads the project. (And the malaria vaccine portfolio is far from robust.)

 “There’s really no silver bullet,” for diagnostics, LaBarre said at a malaria forum in December. “[T]here are many use scenarios and probably no one tool is going to fit all the needs in the way that RDTs have been really instrumental in a one-size-fits-all for control recently.” The costs of RDTs have been driven down because they are one size fits all. If, for eradication purposes, multiple diagnostic technologies move forward, those likely won’t benefit from the same cost-reducing scale of demand. LaBarre, however, does plan “some market shaping to make sure that we can try to achieve the same economies of scale.”

Diagnostics limbo: how low do you need to go?

Malaria’s life-cycle creates a perhaps insuperable detection problem because of a phenomenon called sequestration. Under some circumstances, all parasites in a person’s body sequester themselves outside the blood stream by binding to the inside of blood vessels, for example. Not only do sequestered parasites evade counter-attack by white blood cells and avoid getting filtered out by the spleen, they can potentially confound tests based on a blood sample drawn at just the wrong time. According to Michael Hegg: “Sequestration is an issue for ALL detection methods” that test for the parasite in blood. Sequestration becomes more likely at low levels of infection. Even PCR can, because of fluctuating numbers of parasites, miss infections if they fall below the technology’s limit of detection (LOD).

No one really knows how many parasites must be in a person for them to be capable of transmitting malaria nor how low PCR or some other diagnostic needs to go. According to a PATH document: “existing data are limited, and there is no universal agreement on an exact threshold LOD [Limit of Detection].” At the Gates Foundation, “We spend a lot of time here discussing ‘what does it mean to be infectious?’ ” according Janice Culpepper, who works on malaria at the foundation. “Clearly if you have tons of parasites, you’re likely to be infectious.” However, for very, very low infection levels, assays might find minimal evidence of malaria but, “if you put a mosquito on some of these people, you will infect that mosquito,” said Culpepper in an interview earlier this year.  “[W]hile you may not see much in the peripheral blood, they may actually sequester into the skin, into the capillary beds and things. So while you say, ‘wow, this person looks negative,’ they’re actually in places were mosquitos would bite you. They’re waiting.”

Consequently, everyone is a suspect. The foundation convened a meeting molecular epidemiologists, said Culpepper, “to talk about how low do we think we need to go in our testing to understand where the infectious reservoir is.” Conclusion: “we’re going to say you’re infectious if we have any evidence of any parasite anywhere. Because we don’t know. We may change that definition over time as we get some data.” Added PATH’s Paul LaBarre more recently, “Ongoing and planned studies are aimed at providing the evidence to drive more alignment on LOD [limit of detection] requirements.”

There is no sufficiently sensitive RDT on the horizon, and it is infeasible to test all potentially infected people using PCR. Even PCR could miss infections. It appears practically and perhaps even theoretically impossible to realize the foundation’s vision of complete detection.

Slippery slope to mass drug administration

The lack of sufficiently sensitive point-of-care diagnostics makes mass drug administration the preferred, superior strategy over screening and treatment. “If an insensitive diagnostic is used,” wrote the Gates-funded researchers at Intellectual Ventures, mass screen and treat campaigns “will fail to eliminate a large portion of the parasite reservoir” because infected individuals will be missed. “[M]ass-screen-and-treat campaigns are much less efficacious than mass drug administrations,” the study concluded. Similarly, an earlier investigation, again backed by the Gates Foundation, also found that “modelling shows that MDA has a more pronounced community effect, as all current diagnostic approaches will miss a proportion of infected individuals.” Conceivably, everyone at risk of harboring a malaria infection must be treated.

“Certainly the mathematical models and recent experiences confirm that MDA can produce a faster and more durable transmission impact than the test and treat strategies,” said Patrick Kachur, chief of the malaria branch at the Centers for Disease Control.” However, Kachur added that mass drug administration “won’t be practical or appealing everywhere.” Other tools will be needed, such as vaccines and ways to control mosquitoes. However, the World Health Assembly recently voted to target a 90% reduction in malaria by 2030, leaving little if any time for the development and deployment of a vaccine. Bednets have had a substantial impact but might have already reached and fallen from their high water mark of effectiveness. Outdoor repellents have been a new research emphasis, but they don’t take the fight to malaria like drugs and vaccines, as would be necessary to achieve eradication.

Billions served

Approximately half of the world's seven billion people are at risk of malaria, according  to WHO, although just 1.2 billion are considered to be at high risk. The Gates Foundation is funding research into more precise estimates of the extent of the population malaria eradication efforts would need to encompass. But the numbers will be large. In Africa alone, the Malaria Atlas Project estimated 722 million people in 43 countries were at risk of malaria from Plasmodium falciparum.

Next: Complete Cure.

[Article modified 10:15 AM 7/10/2015]

Vaccine-derived polio case in Nigeria puts eradication milestone in question

A confirmed case of vaccine-derived polio in Nigeria greatly complicates global plans to retire the trivalent vaccine next year and switch to the bivalent formulation. The polio eradication program is now between rock and hard place, with logistical momentum building for the switch but a possible public health emergency should the switch go ahead as planned.

In rare instances, the live oral vaccine can mutate, circulate and paralyze like its former self. Most cases of circulating vaccine-derived poliovirus (cVDPV) are caused by the type 2 virus in the trivalent vaccine, scheduled for retirement in April 2016. But the type 2 component of the vaccine both causes and protects against cVDPVs. In a Catch-22, the trivalent vaccine can’t be withdrawn until it stops the problem it is causing. Pulling the vaccine before halting type 2 cVDPVs would lead to a growing immunity gap and create the conditions for potentially large outbreaks.

Prior to the Nigerian case of cVDPV reported last week, Pakistan had caused the greatest concern with recent sewage samples testing positive for cVDPV. Nonetheless, the World Health Organization confirmed in April the scheduled replacement of trivalent vaccine with bivalent set for April 2016. The bivalent vaccine immunizes against only types 1 and 3 of the poliovirus. Type 2 appears to be long gone, last seen in India in 1999.

The logistics of the switch are daunting: 156 countries currently using or stockpiling the trivalent vaccine need to stop and switch to bivalent at the same time. Every dose of trivalent vaccine administered afterwards creates the risk of type 2 vaccine-derived virus.

In addition to the heavy logistical burden in the field, the switch also requires coordination among manufacturers who must scale back and eventually stop making the trivalent formulation and ramp up bivalent production. Once on, the switch is difficult to turn off.

"An absolute prerequisite"

Until recently, extinguishing all circulating vaccine-derived viruses was an unambiguous precondition for the switch. The eradication endgame plan states that “validation of the elimination of persistent cVDPV type 2…” must precede withdrawal of the trivalent vaccine. The US Centers for Disease Control (CDC) concurred that “persistent cVDPV2s need to be eliminated before the withdrawal of tOPV [trivalent vaccine].” Earlier this year, Paul Rutter, spokesperson for polio eradication's Independent Monitoring Board, said: "My understanding is that the switch could not happen unless cVDPVs are stopped—it is an absolute prerequisite."

No longer.

WHO’s Strategic Advisory Group of Experts (SAGE) decides vaccine policy. SAGE will meet again in October. “The SAGE is not only going to look at whether there is circulation,” said WHO spokesperson, Sona Bari, in early June. According to Bari, SAGE will also consider "what steps have been taken to stop circulation, what immunity levels are like, etc.” 

The Independent Monitoring Board (IMB) backed off from its earlier more absolute position after SAGE gave its go ahead for the switch. Said IMB spokesperson, Paul Rutter: “making a judgement about what constitutes a 'showstopper' would be to second-guess SAGE."

Earlier this year, a modelling study warned of a worrying possibility that vaccine derived virus would still be circulating next year when the switch is set to occur. A co-author of the study, Kimberly Thompson, expressed concern back in February that "It's possible that world leaders will decide to coordinate OPV2 cessation in April 2016 without being 95% confident that cVDPV2 transmission has stopped in Nigeria or Pakistan." At the time, Thompson believed “Pakistan may be more of a threat to global cessation than Nigeria." And subsequently, immunization efforts in Nigeria included measures to drive down cVDPV risk, particularly by vaccinating with the trivalent vaccine. As recently as June 22, Thompson believed Nigeria “can be OK in April 2016 at the time of the switch.”

After the Nigerian cVDPV case last week, however, Thompson stated that "if global health leaders want at least 95% confidence that cVDPV2 transmission has stopped in Nigeria prior to coordinated OPV2 cessation they will need to delay cessation beyond April 2016." Polio’s annual infection cycle is at its low ebb in the month of April. Consequently, a delay in the switch would likely push the date a full year to April 2017.

Pakistan too remains a risk for having cVPDV come next April, according to Thompson, although the risk in both Pakistan and Nigeria can be reduced by the number and quality of vaccination campaigns using the trivalent vaccine.

Thompson and co-authors at the CDC said in a recent paper that switching to bivalent vaccine while vaccine-derived virus circulated “would represent a public health emergency…” WHO already declared polio a Public Health Emergency of International Concern (PHEIC), back in May 2014. The CDC raised polio to a maximal, Level 1 crisis in 2011.

Regarding the schedule for the switch, the Gates Foundation deferred to SAGE. Said foundation spokesperson, Rachel Lonsdale, “The SAGE will review the plans for the switch this fall and make the decision if it is moving forward next year.”

A WHO spokesperson made no comment to an emailed request.

[Article updated at 11:33 am and 11:52 am 7/6/2015]

[Article updated at 3:33 am 7/7/2015]

Has Gavi moved the needle in vaccinating the world's children?

Gavi, long above the fray and untouched by vituperations over aid effectiveness, has come under scrutiny in a paper from the Center for Global Development (CGD). The paper found Gavi “failed to increase vaccination rates for diseases covered by cheap, existing vaccines” and did little better with new, more expensive vaccines. The study came with important caveats but, together with findings of other researchers, jeopardizes not just Gavi’s gold standard status but challenges its—indirect—claim to have saved millions of lives.

Gavi is the poster child of global health. As Bill Gates said recently: “Every time people look at aid money and say what’s the most impactful thing they can find in that whole space, Gavi gets ranked at the very top.” Gavi began as the Bill & Melinda Gates Childhood Immunization Program, and the “work that the foundation has done in vaccines through Gavi,” continued Gates, “has been the most important thing we’ve done.” As a result, “we’ve saved millions of lives,” seven million, according to Gavi, since it began in 2000.

The CGD’s Justin Sandefur agrees that “vaccines are saving millions of lives—I have no doubt about that.” But Sandefur and colleagues examined, “the causal effect of giving Gavi funding in terms of increasing number of lives saved…” He notes that even Gavi does not claim causality but to have helped developing countries save millions of lives. “If you read very carefully, it doesn’t make a clear, causal attribution to Gavi funding doing that,” said Sandefur.  But, “I don’t think that would be a lay person’s interpretation of what’s on the website necessarily.” Readers—and perhaps funders—infer causality. Bill Gates does.

But it is actually a hypothesis that, over the last 15 years and with a budget of $7 billion, Gavi increased the number of children vaccinated, thereby saving millions of lives. “Certainly we can’t corroborate the full headline numbers,” said Sandefur. For the classical vaccines, DPT (Diphtheria, Pertussis, Tetanus) and Hepatitis B vaccines, “we’re not finding any impact,” said Sandefur. CGD found immunization rates were a function of national income, not Gavi support. The CGD study compared countries near the eligibility threshold for Gavi, national income of $1,000 per person. If Gavi raised vaccination rates, there should be a jump right at the eligibility cut-off, because of the sudden jump in available funding.

But there is no jump for the older, inexpensive vaccines, DPT and Hepatitis B. Coverage for those Gavi-supported vaccines looked the same as for measles, which Gavi did not fund.

Gavi spokesperson, Rob Kelly, countered that “The paper misinterprets the statement that Gavi has immunized an additional 440 million children to mean that those children would not have been immunized at all in the absence of Gavi support.” Kelly acknowledged, that "many of these children would have been reached in the absence of Gavi support, but with fewer vaccines.”

Rolling out newer, more expensive vaccines has been a major emphasis for Gavi, both to reduce inequities (it used to take decades for new vaccines to reach the developing world) and to further reduce vaccine-preventable deaths from diarrhea, for example. According to Kelly, “The data makes it clear that the very large majority of Gavi supported vaccines were not in country programs prior to Gavi support.” 

However, CGD found only “small and statistically insignificant effects for the three high-priced vaccines promoted by Gavi,” Haemophilus influenza type B, pneumococcal disease, and rotavirus. According to Sandefur, this is because “…the rotavirus and pneumo vaccines simply haven’t had that wide an adoption in the lowest income countries—yet.” Sandefur said the absence of Gavi’s impact on vaccination for the older, cheaper vaccines is “probably the main thing people are going to take away from the paper,” rather than the “non-robust, positive results on the newer vaccines.”

Fungibility: how even free didn't work

The countries receiving Gavi support seem to spend the money elsewhere: “subsidized vaccines appear to have displaced domestic vaccination spending in countries near the threshold,” concluded the CGD study. The policy implications, said Sandefur, are that “delivering cheap, already existing vaccines to these middle-income countries with millions and millions of poor people is probably not going to accomplish much.” Such countries include India, China and Ghana, for example and “account for a huge share of the global poor.” They also account for a huge share of the lives Gavi claims to save.

The fungibility of aid and vaccine support in particular are not new issues. A Gavi-funded study raised the concern of whether Gavi’s Immunization Support Services (ISS) program displaced national government vaccine spending. Results of the 2007 final report were “inconclusive.” But it said ISS was “the easiest source of funding to meet [national immunization program] priorities…” Consequently, the report recommended more effort to ensure that ISS “does not become the funding source of first resort.” The analysis found worrying signs. Without ISS, spending on activities most likely to raise vaccination rates fell in 20 of 27 countries examined, by a median 4%. Although the changes were not statistically significant, it was “possible that ISS funding has displaced other sources of funding,” a result confirmed by the more recent CGD study.

"Near" the cutoff: The slippery slope to no impact

What about much poorer countries not near the threshold? Sandefur said he is “agnostic” about the impact of Gavi on much lower income states like Niger and Burkina Faso. “Our results are focused on countries near the threshold.” However, for the CGD analysis, “near the threshold” actually means income levels down to $500 a year per person. Thus the CGD findings of no impact technically extend to Gavi-supported, low-income countries like Chad, Somalia and North Korea.

What about still poorer countries? Sandefur cautioned that, “Philosophically, there's also an argument to be made that the causal effect we estimate here applies strictly to a hypothetical country right on the threshold, just teetering on the edge.” This would limit the comparison to Gavi-supported Ukraine, on one side of the cut-off, and the Philippines on the other. However, there's no strict rule about how far to extrapolate these causal effects. “The more you extrapolate, the more damning our conclusions,” said Sandefur. “I would go beyond Ukraine, but stop short of the whole world.” Sandefur noted, however, that methodological objections are not raised to desired results, only negative ones: “if we had found great, positive effects, everybody would be eager to extrapolate to all countries.”

Sandefur acknowledged there is a basis for extrapolating the findings of no impact from Gavi even down to very poor countries. “If there was a big switch in the impact, you would expect to see some big jump in the vaccination rates between lower income and middle income and what I’m saying is you don’t really see that. There’s not some sudden regime shift when you move from low income to middle income countries.” In addition, vaccination rates before and after Gavi started in 2000 can be compared to “see if you see any jumps at that point,” after Gavi kicks in. “And you don’t,” said Sandefur.

A major methodological problem for measuring Gavi’s effect in poor countries is that there is no control group. According to Sandefur: “All very poor countries either got Gavi aid, or were eligible to get Gavi aid and didn’t for some reason that probably implies they're a bad control group…” Absence of a control group has led to “fairly wild extrapolations from what's happening in non-poor, non-Gavi countries,” explained Sandefur.

A 2008 paper from the Institute for Health Metrics and Evaluation (IHME) estimated 7.4 million additional children were vaccinated since Gavi came into being. But Sandefur questioned “whether there are really strong grounds for causal inference or attribution to Gavi on the basis of those regressions, to be blunt.” And indeed, the IHME paper does not assert causality, instead reporting that “we have not addressed whether the increases in DTP3 coverage that have occurred in GAVI ISS recipient countries would have occurred in the absence of GAVI’s support.” 

Gavi support did lead to an increase in reported vaccination coverage. Countries received rewards for increasing vaccination rates and reported rates duly rose, as high as 80–100% in countries like Niger and Mali where surveys put coverage much lower, closer to 40–60%. Whether a positive impact of Gavi would remain absent over-reporting was “an unanswered question,” according to IHME. 

CGD confirmed IHME’s 2008 results. “There is clearly evidence that there was manipulation of the DPT data by Gavi-eligible, ISS-eligible countries,” said Sandefur: “So progress there seemed to have been exaggerated.” Yet, even with the benefit of immunization over-reporting, CGD still found no impact from Gavi.

An undesirable conclusion

Gavi was started because it seemed to many that the urgency had gone out of vaccination efforts led by the World Health Organization and UNICEF. But IHME’s 2008 study found that perception was mistaken: “Globally, survey-based coverage shows that the yearly increase is fairly constant before and after the establishment of Gavi…” Few would question whether Gavi was a good idea. But the intrepid Gates-funded researchers at IHME wondered: “Would Gavi have been implemented if there was less of a perception that improvements in DTP3 immunization coverage during the 1990s were stagnating?” Melinda Gates continues to believe in the now-disproven stagnation hypothesis. Speaking last month of the Gates Foundation’s entry into vaccination, she said: “We started to study it. We started to learn how we had this incredible vaccine system that had kind of crumbled over time.” 

Melinda explained that “we” meant more than just the Gates Foundation. “When I say ‘we,’ I don’t want people to think we’re taking credit for all this,” pointing to the importance of partners. However, the Gates Foundation annual letter laid indirect claim to reducing child mortality: “The last time we cut the child death rate in half, it took 25 years. We will do it again in 15 years.” It’s a story everyone wants to believe. Gavi is a “game-changer,” gushed Melinda Gates’ interviewer. Yet quite possibly, even the very poorest countries needed no help and “we,” however defined, contributed nothing to saving the lives of children, only the perception of saving millions of lives.

Gavi subsumed the Expanded Program on Immunization (EPI), officially begun in 1974 and run by UNICEF and WHO. EPI hoisted vaccination rates from as low as 5% all the way to a reported 80%, a goal declared as achieved in 1990. Today, with Gavi in charge, WHO recently said immunization “stalled in recent years”—the needle stuck at the familiar 80% level.

Gavi recently completed a new $7.5 billion round of financing and promises to save six million more lives.

Have we already seen the best malaria vaccine prospects?

For half a century, scientists century have known one weird trick for eliciting immunity to malaria:  allow a person to be bitten 1,000 times by mosquitoes that have been irradiated. It works. In 1967, researchers reported sterilizing protection in mice from this approach. One thousand bites worked in later human experiments too, eliciting protection in 90% of the small group tested. Decades of malaria research has focused on bottling up the irradiation magic into a vaccine. But the most promising leads have been tried without success, raising the question: is the best yet to come? Or are we scraping the bottom of the barrel?

Malaria infection starts with the bite of a malaria-carrying mosquito which injects malaria spores (“sporozoites”) into humans. Radiation-damaged sporozoites, however, prompt a highly protective immune response while being too weak to survive and develop into disease.

A sporozoite consists of more than one thousand response-provoking, antibody generators or “antigens.” In the 1980s, scientists examined the successful antibody responses to the irradiated sporozoites. The winner, going away, was CSP or circumsporozoite protein. Other antigens were discovered too, with acronyms like TRAP and LSA-1. But none dominated the immune response like CSP, making it the clear favorite for a vaccine that only presented the parts of the parasite needed to elicit immunity, a “subunit” vaccine.

The CSP gene was sequenced, also way back in the 1980s, and optimism ran high that a vaccine was within reach. Much lengthy and laborious research eventually produced a candidate vaccine based on CSP called RTS,S. It was “the logical extension of more than a decade of research built on the hypothesis that a subunit vaccine based on the CS protein would protect humans from malaria infection,” as the vaccine’s co-inventor, W.R. Ballou, put it.

In 1997, RTS,S protected a remarkable 6 out of 7 people. However, fast forward years and hundreds of millions of dollars to 2014, and field trials now show the once-promising RTS,S protects only a disappointing 28% of children vaccinated. Although based on CSP, the champion of antigens, RTS,S delivered less than a third of the 90% benefit promised by the whole organism approach.

Many different approaches, combinations, formulations and technologies have been tried but none have produced a vaccine as good as even the mediocre RTS,S. Malaria researchers have helped pioneer a number of different technologies like the use of viruses and DNA as delivery vehicles. Adjuvants, used to amplify immune responses, have also been varied and are still being tested to see if they can improve the protective benefit of different vaccine candidates.

However, in recent years, little new ground has been broken. “A lot of people have been focused on various formulations with the currently available candidates,” said Lee Hall, Chief of the Parasitology & International Programs Branch at the National Institutes of Health. “There was some ramp up as people looked at a number of different ones. But if they didn’t pan out in the early studies then they weren’t moved forward.”

Among alternative to CSP, only one antigen, based on TRAP, is now in Phase IIb clinical trials. The vaccine protected only 13% of vaccinees in an earlier study. There is one other single-antigen candidate, FMP012, that has pending results from a Phase I trial. Mainly, however, vaccines aimed at protecting against malaria infection are overwhelmingly variations on CSP, both in the PATH Malaria Vaccine Initiative pipeline and the World Health Organization “Rainbow” table. Some projects combine RTS,S with other vaccines, in search of additive or synergistic effect. Also, scientists recently succeed in making a full-length CSP for study. (RTS,S only uses a portion of CSP.)

Many combinations are possible and the number of antigens could be increased, although technical and regulatory issues begin to arise. “Obviously people want to go for something is likely to be successful,” said Lee Hall. “They don’t want to embark on something that is so complex at the outset that the chances of it making it through the development process is low.” How many antigens a subunit vaccine could deliver is not known. According to Hall: “If it turns out that it’s two or five, that might [be] possible to do. If it turns out you need 50, then it turns out to be a lot more difficult.”

Some scientists see the failure of RTS,S as the undoing of the subunit approach. The success of the irradiated sporozoites served “as justification for the ensuing decades of research aimed at identifying the ‘right’ vaccine antigen,” or antigens wrote Mahamadou Thera and Christopher Plowe. “[I]t is reasonable to believe that it may be possible to construct a multistage, multi-antigen recombinant protein that improves on the efficacy of RTS,S,” they continue. But Plowe is skeptical that high levels of protection are possible. In a book chapter version of the article, Plowe added: “it seems not unlikely that vaccines that target 2, 5, or even 15 of the 5,000 gene products will still fall short of the high levels of protection seen with radiation-attenuated whole-organism vaccines when delivered through the bites of infected mosquitoes.”

According to Timothy Wells, Chief Scientific Officer at the Medicines for Malaria Venture, RTSS “even as a partial vaccine, is a massive triumph immunologically.” The human immune response to malaria is transient and requiring, on some estimates, 10 malaria infections, according to Wells. “This is why making a vaccine is close to impossible,” said Wells. Perhaps informed by this conclusion, the PATH Malaria Vaccine Initiative (MVI), funded by the Gates Foundation, has been quietly dropping candidate after candidate. Two projects testing multiple antigens have been dropped, one as long ago as December. However, that cancelation was only recently reflected in MVI’s portfolio which “was just updated last week,” the week of May 10, according to PATH spokesperson, Kelsey Mertes. One other canceled vaccine project “is still included on the portfolio because it hasn’t been closed out yet,” according to Mertes. Cancelled projects remain in the portfolio “until the contract is closed out and the last payment made,” Mertes explained. The portfolio still contains two studies of RTS,S given with another vaccine. Results are in for one but, perhaps suggesting the outcomes were not spectacular, Mertes did not reply to an email asking if it would move forward into the next phase of trials. The MVI portfolio looks increasingly barren as the sole remaining active trial for a protective pre-erythrocytic vaccine will report results later this year.

The search continues technically, although at the very earliest stage of the pipeline with any actual vaccine many years away. According to Mertes, MVI “is undertaking target validation work” on over 25 candidate antigens for a protective, pre-erythrocytic vaccine. However, science suggests that future discovery of a highly protective antigen or antigens is unlikely. Malaria has some 5,000 genes, but there are not 5,000 potential vaccine targets. Malaria genes produce just under 2,000 proteins. Vaccine targets are fewer still because not all the proteins are seen by the human immune system. Indeed, according to Stefan Kappe: “The number of surface and secreted proteins we identified is much lower, less than 100.” Kappe leads a group at Seattle Biomed that began assessing new antigen candidates in late 2012 for MVI. The project is expected to take six years or more. “Most of the novel targets are not published yet,” said Kappe.

It is not the first such effort but, because of its comprehensiveness, it could be the last.

In 2011, a group at New York University published results of efforts, supported by the Gates Foundation and others, to find new candidate antigens for a malaria vaccine. The researchers tested 34 new antigens, but concluded: “In summary, we failed in our attempt to discover powerful protective non-CS antigens…” Researchers saw no reason to look any further for other antigens. The study’s lead author, Satish Mishra, wrote in email: “We can’t rule out the possibilities in science but we as far as we understand it’s very unlikely” that there are important, yet to-be-discovered antigens. More bluntly, according to Mishra: “We have already checked the best candidates and already given up the new candidate search.”

Claim of Possible Drug-Resistant Malaria in Angola Withdrawn

A paper reporting a possible case of drug-resistant malaria in Angola appeared last July (reported here), meeting with scathing incredulity and direct calls for retraction from the World Health Organization and some members of the malaria research community.

The surprising firefight appears to have been concluded, not with retraction but the publication of letters from the authors of the original paper and its detractors at WHO and Mahidol University. The paper’s authors now acknowledge that “our report of a Vietnamese worker returning from Angola with severe Plasmodium falciparum malaria not responsive to artemisinins is unlikely to indicate that artemisinin resistance has reached Angola.”

The paper and letters appeared in Emerging Infectious Diseases, published by the Centers for Disease Control (CDC).

Critics point out that the delayed clearance time seen in the case was nearly ten times slower than that seen in drug-resistant malaria in Southeast Asia, the only region in the world with clinically-verified resistance to artemisinin-based anti-malarials. Genetic testing after the publication of the paper did not find any of the mutations in the K13 gene associated with resistance. Also, the drug used in the case came from a batch that was withdrawn because of quality issues.

In rebuttal, the authors of the original paper note that the drug used, while substandard was far from lacking any activity. The paper is not being withdrawn and the case is not closed, say the authors: “the reasons for the lack of response to artemisinins in this patient remain unknown and are under continued investigation.”

WHO also says it is still looking into the same geographic area of Angola from which the disputed case originated.

Olivo Miotto, who previously predicted the paper would be retracted, said more recently “it was a mess-up, as was suspected at the time by those more interested in science than fiction.”

According to Annette Erhart, one of the paper’s authors, the Vietnamese investigators knew that publishing the case would expose them “to criticisms about the way the patient was managed.” The group published nonetheless because “for the first time, they were confronted to a situation where AS [artesunate] did not work at the dosage recommended by the national guidelines.” Continued Erhart: “The attitude of the WHO is discouraging anybody to report suspected resistance cases, while it should be actually the opposite.”

WHO’s Pascal Ringwald, contacted last September about the case, said “Globally it is amazing how journalists misunderstand artemisinin resistance and they do not investigate on what should be investigated.” Although the paper had been published by the CDC and elicited public statements of concern from prominent malaria researchers, Ringwald insisted on his own knowledge: “I do not know which is or are your sources but I think that you do not trust me.” The case, he said, "is absolutely not interesting and not important.” Ringwald forbade all further inquiries: “In the future please use your other sources of info and do not contact me for any kind of reason.”

A ceasefire of sorts appears to have been brokered. Said Miotto: “I think everyone is just letting it drop—not worth any arguments.” Patrick Kachur, chief of malaria at the CDC, declined to answer questions about the matter directly, referring inquiries to the CDC press office. Kachur did allow that “I was hoping these [letters] would be ready to share a lot sooner.”