The endpoint that declared success
INTerpath-001 assigned 1,137 adults with completely resected stage IIB-IV cutaneous melanoma 2:1 to intismeran autogene plus Keytruda or to Keytruda alone, dosing the vaccine every three weeks for up to nine doses across roughly a year. Merck and Moderna report that recurrence-free survival, the primary endpoint, and distant metastasis-free survival, the key secondary endpoint, were both met, with no new safety signals. That is the first Phase 3 result an individualised neoantigen therapy has produced.[2]
What the announcement does not carry is the size of the difference. The topline holds no hazard ratio, no confidence interval and no median follow-up, and overall survival is still accruing while the trial continues. The comparator was Keytruda alone, which is the right comparison for this group of patients, so the fairness of the match is settled. The open question is magnitude and durability.[2]
What 173 months of follow-up shows
The other number came from 16 SEER registries: 20,057 adults diagnosed with Hodgkin lymphoma between 2000 and 2017 who were still alive 5 years later. Followed for a median of 173 months, they produced 2,899 deaths against 1,480.8 expected in the general population, a standardised mortality ratio of 1.96 with a 95 percent confidence interval of 1.89 to 2.03 and 81.9 excess deaths per 10,000 patient-years. Hodgkin lymphoma is one of the cancers medicine genuinely cures; this is the arithmetic behind that cure once the follow-up runs long.[1]
The causes matter more than the ratio. At 10 years cumulative mortality is 8.0 percent: 2.5 percent from noncancer causes, 2.1 percent from other cancers, 1.6 percent from cardiovascular disease and 1.6 percent from the lymphoma itself. The relative excess is largest in people diagnosed between ages 20 and 39, the ages at which this disease is common, while the absolute excess is largest in the oldest group, at 346.1 per 10,000 patient-years. SEER holds no clinical or socioeconomic detail, so the authors state plainly that their radiotherapy comparison cannot be read causally, and that follow-up may still be too short for late radiation effects.[1]
Where the two numbers meet
Both results are bounded by the same thing: the length of the clock. A trial of post-surgical treatment has to report on a horizon short enough to be useful, so it reports recurrence and distant metastasis, and it reports them while overall survival is still accumulating. The Hodgkin cohort shows what the far end of that clock looks like in a disease whose short-horizon endpoints were met decades ago, and the deaths that accrued there sit mostly outside the lymphoma, in noncancer causes, second cancers and the heart. A reader could reasonably object that the two say nothing about each other, since Hodgkin survivors carry the late toxicity of an older era of chemotherapy and radiotherapy, while a vaccine plus pembrolizumab is a different exposure with its own late-effect profile. That objection is fair, and it is exactly why the late profile of this combination counts as unknown rather than reassuring.[1], [2]
Three days ago this column argued, over the DECISION trial, that a headline percentage means as much as the analysis carrying it can support and no more. The same discipline applies to an endpoint: recurrence-free survival supports a claim about recurrence inside the follow-up window, and it does not yet support a claim about how long these patients live. So, a testable statement: when Merck and Moderna present INTerpath-001 at the international medical meeting they have promised, the recurrence-free survival hazard ratio will be reported while overall survival is still immature. The signal worth watching is whether the presented data carry a median follow-up long enough for the distant metastasis-free survival curves to separate and stay separated. That curve, more than the endpoint's name, is what tells a patient anything.[2], [3]