High profile cancer trial is withdrawn
Nature has retracted a much discussed study that suggested the timing of immunotherapy could dramatically affect outcomes for patients with advanced non-small cell lung cancer.
The paper, published on Feb. 2 in Nature Medicine, attracted immediate attention across the biomedical community because it appeared to show that patients treated earlier in the day had substantially better survival outcomes than those treated after 3 p.m.
By Feb. 19, the journal had already added an editor’s note warning readers about concerns related to the trial. After a four month investigation, the journal decided to withdraw the article.
Editors lose confidence in the results
The retraction note stated that, because of the volume and nature of the problems identified, the editors no longer trusted the integrity of the findings.
The study, titled “Time-of-day immunotherapy in non-small cell lung cancer: a randomized phase 3 trial,” had been presented as a randomized phase 3 trial known as LungTIME-C01.
The trial was conducted in China and enrolled 210 randomized patients with advanced non-small cell lung cancer.
The 3 p.m. finding that drew attention
The study claimed that patients receiving their first four cycles of a PD-1 inhibitor plus chemotherapy before 3 p.m. had far better outcomes than those treated later in the day.
The immunotherapy drugs used in the trial were Merck & Co.’s Keytruda, also known as pembrolizumab, and Innovent Biologics’ Tyvyt, also known as sintilimab.
According to the now retracted paper, the early treatment group had a 60% lower risk of disease progression or death. Median progression free survival was reported at 11.3 months before 3 p.m., compared with 5.7 months after that time.
Overall survival also appeared to favor earlier treatment. The study reported a 58% lower risk of death, with median overall survival of 28 months in the early treatment group versus 16.8 months in the later treatment group. Both progression free survival and overall survival were reported with p values below 0.001.
Experts were cautious from the beginning
The size of the reported benefit prompted enthusiasm but also immediate caution. Charu Aggarwal, M.D., who leads the head and neck and thoracic cancer practice at Penn Medicine, described the magnitude of the effect as surprising and said it required careful evaluation.
Aggarwal emphasized the need for international, multicenter prospective studies to test whether the finding could hold up across broader patient populations and with stronger control of potential confounding factors.
Daina Graybosch, Ph.D., an analyst at Leerink Partners, viewed the work as hypothesis generating because it came from a single center, Hunan Cancer Hospital in China. Still, she said its prospective and randomized structure, combined with the large reported effect, made the finding difficult to ignore before the investigation began.
Timing question had a biological basis
The idea behind the study was not without scientific rationale. Research in circadian immunology has suggested that immune activity changes over the course of the day, with certain immune cells showing stronger activity during daylight hours.
The retracted paper also reported that early day treatment was associated with higher levels of circulating CD8+ T cells and a higher ratio of activated to exhausted CD8+ T cells. The authors suggested this might help explain the clinical outcomes.
However, researchers also noted that the data could not clearly separate correlation from causation.
Other studies had pointed in a similar direction
Several retrospective studies have explored whether treatment timing may matter in cancer care, although such analyses can be affected by hidden differences among patients.
An international retrospective study of more than 1,000 patients with relapsed or refractory large B-cell lymphoma linked morning infusion of CD19 directed CAR-T therapy with a higher one year progression free survival rate than afternoon treatment. Overall survival, however, was similar. That study was published in Blood in March 2026.
Even so, many experts remained skeptical that a single infusion time could create major clinical differences, especially because PD-1 antibodies and CAR-T cells remain active in the body long after administration. Keytruda has a half-life of more than 20 days.
Questions emerge around trial changes
Several concerns focused on the trial record and protocol history. The ClinicalTrials.gov entry showed major changes after the study began in 2022.
Those changes included shifting the study type between interventional and observational in 2024, modifying primary and secondary endpoints, and substantially revising enrollment criteria.
Although such changes do not automatically invalidate a trial, they can make interpretation more difficult and raise questions about potential bias.
The protocol submitted with the Nature Medicine article was dated Jan. 2, 2022, yet it included references to studies published later in 2022, 2023 and 2024.
Protocol discrepancies add to concerns
Nature’s review found differences between the original Chinese protocol and translated versions supplied to the journal during peer review and publication.
The authors later provided a time stamped version of the original protocol and attributed the discrepancies mainly to administrative errors rather than formally documented protocol amendments.
That explanation did not resolve the broader concerns, especially when combined with unusual patterns in the trial data.
Survival curves and follow-up patterns questioned
One issue involved the progression free survival curve, which appeared smoother than expected for a prospective oncology trial.
In many cancer trials, imaging assessments occur at fixed intervals, often creating survival curves that drop in more visible steps. In LungTIME-C01, tumor assessments were supposed to occur every two 21 day treatment cycles.
The authors told Nature that COVID-19 disrupted fixed calendar imaging schedules and cycle based assessments, which they said affected the pattern of the data.
Safety data also drew scrutiny
The trial reported no censoring during the first year, meaning every patient was successfully followed during that period. It also reported no adverse events that led to treatment discontinuation in either study group.
Those findings were unusual, although Aggarwal said proactive toxicity management can allow many patients to continue therapy without needing to stop treatment.
For comparison, Akeso’s Harmoni-6 phase 3 trial in China, which studied BeOne Medicines’ PD-1 inhibitor Tevimbra with chemotherapy in first line squamous non-small cell lung cancer, reported treatment related adverse events leading to discontinuation in 4.5% of patients. That trial was about 2.5 times the size of LungTIME-C01.
Randomization timing becomes another red flag
Nature also identified concerns in the source data provided by the research team. The journal found that randomization was performed on the day of treatment for almost all patients, a pattern described as uncommon.
The investigation also cited problems related to the timing of tumor progression analysis and the lack of adherence to standard assessment schedules.
Taken together, the protocol issues, data patterns and timing concerns led the editors to conclude that the results could no longer be relied upon.
Authors agree or do not respond
According to the retraction note, the authors either agreed with the withdrawal, including all corresponding authors, or did not respond to correspondence from Nature.
The research team also did not respond to a request for comment.
The retraction leaves the broader question of treatment timing unresolved. Circadian biology remains scientifically plausible, but the withdrawn LungTIME-C01 paper can no longer serve as reliable evidence that giving PD-1 based immunochemotherapy before 3 p.m. improves outcomes in advanced non-small cell lung cancer.
