Criticism of Surrogate Endpoints in Whose Interests?

Last week, in an op-ed in the Wall Street Journal, Mark Thornton, lit into an inquiry launched by US Senator Charles Grassley (R-Iowa). Dr Thornton's concern was that Grassley was attacking the use of surrogate endpoints in clinical trials of drugs used to treat life-threatening conditions. Thornton was described as a "former medical officer in the Office of Oncology Products at the Food and Drug Administration," who also volunteers as "president of the Sarcoma Foundation of America."

Before discussing Dr Thornton's points, let me take a brief detour to explain surrogate endpoints. Ideally, physicians ought to prescribe treatments to alleviate pain and suffering, improve function, and delay death. So ideally treatments ought to be assessed on the degree they affect such endpoints. Often, however, trials designed to measure such endpoints are difficult, and often require prolonged observation of multiple patients. Thus, surrogate endpoints, which are those thought to correlate with such clinical endpoints, are studied instead.

Here is what Dr Thornton wrote about such endpoints:

At issue is the concept of 'surrogate endpoints' and the FDA's 'accelerated approval' regulations.

New laws and regulations ... created an accelerated approval mechanism by which a drug could be allowed on the market if it showed early evidence of an effect on a surrogate endpoint. For cancer, examples of surrogate endpoints are tumor shrinkage or a delay in the disease's progression.

This kind of measurement – as opposed to an assessment of a drug's impact on a patient's overall survival – has dramatically increased the pace of cancer clinical trials. It also has won near-universal acceptance within the cancer community.

The FDA does require follow-on studies to assure that a surrogate finding shows clinical benefit. But if all cancer clinical trials were required to show a survival benefit from the get-go, progress in cancer-drug development would slow to an absolute crawl.

The damage done by Mr. Grassley's decision to make an issue of this decision cannot be understated.

U.S. cancer-drug development stands on a precipice overlooking a new dark age in which each new product's development is longer and costlier than the last. Companies may decide it is not financially viable to even bother developing new drugs, and the pipeline for new products to treat cancer could slow even more. Mr. Grassley's legacy could be thousands of additional cancer deaths.

I cannot comment on the political or bureaucratic issues raised.

But I do question Dr Thornton's uncritical acceptance of surrogate endpoints, and his assertion that any questioning of the use of such endpoints could lead to death and disaster.

The problem with surrogate endpoints is that they are surrogates for the real thing. In many cases, a treatment may appear beneficial when measured by its affect on such endpoints, but not turn out to be beneficial when measured by its affect on real clinical outcomes, e.g., alleviation of symptoms, improvement of function, and prolongation of survival. There are many reasons why this may be the case.

Consider, for example, cancer "progression-free surival," cited by Dr Thornton in his article. Progression-free survival means the time a patient survives without measurable evidence that his or her tumor has gotten worse. There are some obvious reasons why progression-free survival may not correlate with overall survival. First, a tumor may progress but its progression may go undetected. Second, although a treatment may inhibit progression for a while, a tumor could escape its effects, and thereafter progress more rapidly than it did before. Third, a treatment might retard tumor growth, but cause severe adverse effects that hasten death on their own.

Thus, the FDA usually requires studies not only of progression free survival, but eventually of overall survival. Approving a drug only on the basis of its effects on progression-free survival risks allowing the use of a drug that may eventually prove to have no effect on overall survival, and hence whose benefits may not outweigh its harms.

So, regardless of the politics involved, Dr Thornton's dire warning that Senator Grassley's inquiry could lead to "thousands of additional cancer deaths" appears way over the top.

Incidentally, soon after the article came out, Ed Silverman on the Pharmalot blog reported that Dr Thornton, in addition to his voluntary post with the Sarcoma Foundation of America, serves as Senior Vice President of Product Development for the biotechnology firm GenVec, which is attempting to develop drugs to treat cancer. Also, the Sarcoma Foundation of America is supported in part by pharmaceutical and biotechnolgy companies including Ariad, Ziopharm, Bristol Myers Squibb, Ortho Biotech, and Novartis.

So readers of Dr Thornton's op-ed on health policy vis a vis how treatments for cancer and other life-threatening diseases are evaluated by the FDA deserved to know that Dr Thornton works for a biotech companies whose products could end up being evaluated by the FDA, and that Dr Thornton leads a not-for-profit organization which appears to be sponsored in part by other drug and biotech companies whose products also are being evaluated by the FDA.

Whether or not these firms' concerns that limitation of the use of surrogate endpoints in drug approval could affect their profts affected Dr Thornton's thinking is unknown. However, this question would not likely have even occurred to readers who did not know about Dr Thornton's ties, direct and indirect, to these firms.

In lieu of such disclosures, Dr Thornton's original op-ed, appears to be a new and prominent example of stealth health policy advocacy.

Health care is in crisis in many ways. Policy-level solutions deserve to be discussed widely by all sorts of people representing all sorts of opinions, with all sorts of expertise and backgrounds. However, the discussion would be a lot more productive if those taking part were completely transparent about who they are, for whom they work, and whose interests they may serve.