Is the WaPo journalistically challenged when it comes to science? You decide.
An April 6 article in the Washington Post, one of the most recent attacks of the NIH as it undergoes administrative changes, claims that “A big step forward in cancer therapy has been slowed by layoffs and new restrictions at the National Institutes of Health, where it was developed.” The focus is on a Nature Medicine article[1] published on April 1 that is presented as a crucial advancement in “utilizing a patient's own immune cells to combat gastrointestinal cancers.” The Washington Post article delves into the study’s background and theory rather than its results, which allows it to conveniently obscure the fact that the therapy is anything but a breakthrough. Distressing instances of continuing patient treatment despite evidence of the therapy's ineffectiveness, alongside an unprecedented 355 modifications to the study protocol over 14 years, reveal serious oversights requiring thorough investigation by the NIH into its intramural research practices.
In oncology, when there are no effective treatments available for patients, researchers often conduct non-comparative studies focusing on one treatment group. To ensure patient safety, Simon's 2-stage study design[2] is employed (the researchers cited this as the design that they were following). The idea is simple: researchers want to see if there's any positive effect, like tumor reduction, from the new therapy while minimizing patient exposure to ineffective or risky treatments. In the first stage, thirty patients are enrolled, aiming for a minimum response rate of 20% to justify further enrollment (indicating a significant improvement compared to leaving patients untreated). If fewer than 20% of patients respond, the study halts immediately to avoid harm to patients without proof of treatment benefit[3], a practice that has been standard since the 1980s.
The purpose of this study was to evaluate the potential of neoantigen-specific tumor-infiltrating lymphocytes (TILs) as therapies for solid tumor cancers, specifically focusing on gastrointestinal cancer. The researchers aimed to compare patient responses between standard TILs and neoantigen-specific TILs (which they called “selected TILs”). While standard TILs showed a negligible clinical response, the novel selected TILs demonstrated a mere 7.7% response rate, falling short of the 20% threshold for a "modest response." These patients showed only partial improvement; no patient saw a complete elimination of their tumor, which only further supports the notion that this therapy’s activity was not significant. To make matters worse, the patients received chemotherapy and FDA-approved biological therapy in addition to the experimental cell therapy, which could, on their own, be entirely responsible for any tumor shrinkage that was observed.
At this point, had the investigators adhered to the principles of their trial design, they would have terminated the study, preventing further patient exposure to a therapy with no clear benefit. However, they proceeded, administering selected TILs along with the commonly used immune checkpoint inhibitor pembrolizumab. The addition of pembrolizumab, which is already known for blocking the PD-1 protein on T-cells and enhancing their activity against cancer cells[4], resulted in a 23.5% patient response rate. The researchers used this to claim that their trial demonstrates that “TILs selected for neoantigen reactivity can mediate tumor regression even in treatment-refractory metastatic GI cancers that are not thought to be sensitive to immunotherapy.”
The logical error here is astounding: the new intervention alone failed to trigger a notable patient response, yet the conclusion was erroneously drawn that the intervention was responsible, disregarding the marked impact of a known-effective drug. In addition to a grave violation of basic clinical reasoning, this study displays alarming disregard for standard clinical trial ethics. As mentioned previously, this “breakthrough” therapy uses a highly toxic conditioning regimen of chemotherapy and high-dose biologic therapy (another FDA-approved biologic therapy called interleukin-2). Not only could individualized responses of patients to chemotherapy significantly influence the effectiveness of the novel selected TIL therapy, but all patients also experienced severe toxicity, with 10% requiring hospitalization. More specifically, “seven patients required escalation of care for critical support: three for mechanical ventilation…one for continuous renal replacement therapy…three required both modalities…” and there was even “one treatment-related death attributed to adenoviral hepatitis 49 d after TIL infusion.”
The Washington Post’s sensationalistic and poorly informed journalism has inadvertently exposed disturbing negligence on the part of NIH investigators. A gross lack of adherence to basic clinical trial study design and ethics resulted in significant harm to patients without any discernible benefits. We call upon the NIH to investigate this trial thoroughly and use its findings to install necessary oversight mechanisms across its clinical research activities.
George F. Tidmarsh, MD, PhD, Stanford University School of Medicine, is an Adjunct Professor, Pediatrics and Neonatology. He is the Founder, M-TRAM, and Senior Advisor, Center for Advanced Pediatric and Perinatal Education.
Arman Sharma, a recent graduate of Stanford University, helped in writing and research for this article.
[1] https://www.nature.com/articles/s41591-025-03627-5
[2] https://brb.nci.nih.gov/techreport/Optimal2-StageDesigns.pdf
[3]https://pmc.ncbi.nlm.nih.gov/articles/PMC6047527/#:~:text=A%20subtle%20….
[4] https://www.cancer.gov/about-cancer/treatment/drugs/pembrolizumab
