The C8 cohort has now been followed for deaths, and brain cancer rose

Almost every claim made in public about the health effects of PFOA traces back, directly or indirectly, to one group of people: residents of the mid-Ohio Valley whose drinking water was contaminated by a fluorochemical plant, and who were enrolled in a cohort of 32,254 highly exposed persons [s1]. That cohort gave the C8 Science Panel its findings — probable links between PFOA and kidney cancer, testicular cancer, raised cholesterol, ulcerative colitis, thyroid disease and pre-eclampsia [s1]. Those findings were based on disease incidence followed through 2011 [s1].

A paper posted on 30 August extends the same cohort by a decade, and changes the question from who got sick to who died [s1].

What was done

Deaths were ascertained through the National Death Index, and the researchers extended their cumulative PFOA serum estimates from 2011 forward to 2021 [s1]. They then used Cox regression with internal comparisons — that is, comparing more-exposed cohort members against less-exposed ones rather than against the general population — to analyse cause-specific mortality for 22 causes of death against the natural log of cumulative PFOA exposure, both as a continuous variable and in quartiles, with a variety of lags and adjustment for confounders [s1]. Multiple cause of death was also considered [s1].

There were 5,709 deaths in the cohort through 2021 [s1].

What came out

One cause of death showed a positive significant trend in the underlying-cause analyses: brain cancer, at a hazard ratio of 1.25 (95% CI 1.05 to 1.50) per log unit of cumulative serum level [s1]. Bladder cancer was suggestive but not statistically significant, at 1.17 (0.98 to 1.39) [s1]. In stratified analysis, pancreatic cancer showed a positive significant trend among participants older than 72, at 1.23 (1.05 to 1.43) [s1].

No other causes showed marked positive trends, with the exception of stroke among workers, who made up 13% of the total cohort [s1].

Why the authors are cautious, and why that matters

The paper's own conclusion carries an explicit warning: caution is warranted due to multiple comparisons [s1]. This is not boilerplate modesty. Testing 22 causes of death, in continuous and quartile form, across a variety of lags, generates a large number of statistical tests. At a conventional threshold, some of them will cross significance by chance alone. A single significant trend out of 22 causes is roughly what chance would produce.

That does not make the brain cancer finding meaningless. It means the finding is a candidate for replication rather than a conclusion. The distinction the study itself draws is between results that arrive with independent support and results that do not. Pancreatic cancer arrives with some: it has been associated with PFOA in animal studies, with what the authors describe as equivocal evidence in other cohort studies [s1]. Brain cancer, the one result that reached significance in the main analysis, does not come with that kind of corroboration named in the paper.

It is also worth being precise about what these hazard ratios describe. They are internal comparisons within an already highly exposed population [s1]. They say how mortality varied with cumulative exposure inside that cohort. They are not a comparison between this community and an unexposed one, and they cannot be read as an absolute risk to anybody.

What did not appear

The negative space in this study deserves as much attention as the positive findings. The C8 Science Panel's probable links included kidney and testicular cancers [s1]. Neither shows up among the mortality trends reported here. That is not the same as a refutation — incidence and mortality are different outcomes, kidney and testicular cancers are relatively survivable, and a mortality study is a blunt instrument for detecting a disease people do not usually die of. But it does mean this decade of follow-up did not simply confirm the earlier picture. It produced a different one.

The wider frame

PFAS comprise thousands of anthropogenic fluorinated compounds whose stability comes from the carbon–fluorine bond, the strongest single bond in organic chemistry — the property behind the "forever chemicals" epithet [s2]. Decades of use in aqueous film-forming foam, industrial processes and consumer products have created persistent, highly mobile groundwater plumes now detected on every inhabited continent [s2]. The first legally binding US drinking water limits, 4 ng/L for PFOA and PFOS, were promulgated in 2024 [s2].

The mid-Ohio Valley cohort is the reason those limits have a health rationale attached to them at all. It is the largest and longest-followed group of highly exposed people anyone has assembled, and the evidence base is thin enough that a single cohort carries this much weight. That is precisely why a result from it should be read carefully rather than loudly: when one study is doing the work of many, an overstated reading of it propagates everywhere.

What to watch is whether the brain cancer signal survives. Cohorts of this kind get re-analysed; other highly exposed populations exist in Europe and Asia; and the authors have flagged bladder cancer as worth another look. Replication, not this paper, will settle it.

Sources

• [s1] Mortality study of a PFOA-exposed cohort in the mid-Ohio Valley. Environment International, 30 August 2026. https://doi.org/10.1016/j.envint.2026.110492
• [s2] From forever to fluoride: A critical review of per- and polyfluoroalkyl substance (PFAS) occurrence, fate, and the transition from sequestration to destruction-based remediation in groundwater. Science of The Total Environment, 31 August 2026. https://doi.org/10.1016/j.scitotenv.2026.182278