How Long Until Immunotherapy Pays Off in Lung Cancer? New Study Puts a Number on It

For patients with operable lung cancer, one of the most practical questions in modern oncology has gone strangely unmeasured: how long does it actually take for immunotherapy to start working? A new comparative effectiveness study published in BMC Cancer has tackled that question with unusual quantitative rigor, and its answer is a near-linear curve of accumulating benefit that reaches clinically meaningful territory at roughly nine to ten months of treatment. The finding, drawn from reconstructed patient-level data across ten randomized controlled trials involving 5,575 participants, offers oncologists something they have rarely had: a timeline for when the promise of perioperative immune checkpoint inhibitors converts into measurable reductions in recurrence and death.

The study, led by Yongyan Lu, Wenming Bian and Xiaohui Jia of Xi’an Jiaotong University, with Lili Jiang, Hui Guo and Chao Li as corresponding authors, addresses a genuine gap in the evidence base. Perioperative immunotherapy—immune checkpoint inhibitors given both before and after surgery for resectable non-small-cell lung cancer, or NSCLC—has transformed treatment over the past several years, delivering striking gains in event-free survival and overall survival across landmark trials. Yet the optimal duration of that therapy has never been formally defined, and conventional efficacy measures such as hazard ratios say little about when, in the course of treatment, the benefit actually begins to accrue.

To fill that gap, the researchers turned to a metric called time to benefit, or TTB. Unlike a hazard ratio, which summarizes the relative difference between two survival curves, TTB quantifies how therapeutic effects accumulate over time. It answers a question patients and clinicians ask constantly: if I commit to this regimen, how many months of it do I need before my absolute risk of recurrence or death drops by a meaningful amount? The measure bridges the space between treatment duration and clinical benefit, and it is particularly relevant for perioperative regimens, where months of therapy carry real costs in toxicity, inconvenience and expense.

The methodology behind the analysis is as notable as the result. Because individual patient data from the ten trials were not directly available, the team reconstructed it by digitizing published Kaplan–Meier survival curves, a well-established technique that extracts approximate time-to-event information from the stepwise plots that randomized trials publish. The trials spanned the major perioperative and adjuvant immunotherapy programs in resectable NSCLC, including AEGEAN, Neotorch, KEYNOTE-671, CheckMate-77T, CheckMate 816, NADIM II, RATIONALE-315, IMpower010, KEYNOTE-091 and TD-FOREKNOW. Supplementary analyses published alongside the paper show side-by-side comparisons of the reconstructed curves against the originals, a transparency step that lets readers judge the fidelity of the reconstruction for themselves.

With the pooled dataset assembled, the researchers applied Weibull survival modeling combined with Monte Carlo simulation to estimate the time required to reach predefined thresholds of absolute risk reduction, or ARR. The choice of Weibull models matters: unlike some parametric approaches, Weibull distributions can capture both increasing and decreasing hazard over time, which is essential when the underlying biology involves an immune response that builds gradually rather than a cytotoxic drug effect that is front-loaded. Monte Carlo simulation then propagated the statistical uncertainty through the estimates, so that each reported time-to-benefit figure reflects the range of plausible values rather than a single brittle point.

The headline numbers are strikingly clean. For event-free survival—the risk of recurrence, progression or death—roughly 1.02 months of perioperative immunotherapy were required for each 1% absolute reduction in risk, and 9.01 months for a 10% reduction. For overall survival, the calculus was slower: a 1% reduction in mortality risk required approximately 10.67 months of therapy. The benefit accumulated in a near-linear fashion, which the authors interpret as evidence that the immune-mediated protection builds steadily across the treatment period rather than arriving in a burst. Taken together, the entire perioperative strategy reached the prespecified thresholds for reducing both recurrence-related events and mortality at approximately nine to ten months.

That timeline carries immediate practical weight. Perioperative immunotherapy regimens in NSCLC typically involve three to four cycles of neoadjuvant therapy before surgery, followed by up to a year of adjuvant treatment. A patient weighing whether to continue adjuvant infusions at month six—perhaps fatigued by side effects or questioning the value of continued treatment—can now be told that the analysis projects the 10% recurrence-risk reduction threshold arriving at around nine months. Conversely, the near-linear accumulation suggests there is no obvious early point at which the bulk of the benefit has already been banked, which cuts against casual early discontinuation.

The subgroup analyses add a layer of clinical nuance. The study found that patients with squamous histology, stage III disease, or PD-L1 expression of at least 1% reached the prespecified absolute risk reduction thresholds earlier than other patients. PD-L1, the programmed death-ligand 1 protein that immune checkpoint inhibitors target, has long served as a rough biomarker for immunotherapy responsiveness, so its appearance in the earlier-benefit group is biologically coherent. Stage III disease, being more advanced, offers more absolute risk to reduce, which mechanically shortens the time needed to hit any fixed ARR threshold. Squamous histology has previously been associated with stronger immunotherapy signals in lung cancer, and this analysis is consistent with that pattern.

The study’s framing of TTB as a complementary measure deserves emphasis. The authors are not proposing that time to benefit replace hazard ratios, median survival or any established endpoint; rather, they position it as an additional lens that reveals the temporal texture of a treatment’s effect. Two therapies can post identical hazard ratios while delivering their benefit on very different schedules—one front-loaded, one gradual—and that difference is invisible to conventional pooled analyses but decisive for real patients deciding how long to persist with treatment. In an era when immunotherapy courses are long and expensive, quantifying the schedule of benefit is arguably as important as quantifying its magnitude.

The work also illustrates a broader methodological trend in oncology: the extraction of individual-patient-level insight from aggregate published data. The Xi’an Jiaotong team, working across the university’s Second Affiliated Hospital, its School of Public Health and partner institutions, obtained ethics approval from the Xi’an Jiaotong University Health Science Centre institutional review board, with the consent requirement waived because the analysis was secondary and based on published results. Funded by the National Natural Science Foundation of China and several Shaanxi provincial and clinical oncology grants, the study demonstrates that carefully validated curve digitization, pooled survival analysis and parametric simulation can squeeze genuinely new clinical guidance out of trials that have already reported their primary results. For the growing population of patients facing surgery for lung cancer, the message distilled from 5,575 reconstructed patient journeys is a concrete one: the benefit of perioperative immunotherapy is real, it accumulates steadily, and the milestones that matter most arrive at around the nine-to-ten-month mark.

Subject of Research: Time to benefit of perioperative immunotherapy in resectable non-small-cell lung cancer

Article Title: Time to benefit of perioperative immunotherapy among patients with resectable non-small-cell lung cancer

Article References: Lu, Y., Bian, W., Jia, X., Li, J., Liu, M., Zan, Y., Fu, X., Liu, Z., Huang, L., Li, C., Guo, H., & Jiang, L. (2026). Time to benefit of perioperative immunotherapy among patients with resectable non-small-cell lung cancer. BMC Cancer. https://doi.org/10.1186/s12885-026-16662-3

Image Credits: AI Generated

DOI: 10.1186/s12885-026-16662-3

Keywords: non-small-cell lung cancer, perioperative immunotherapy, immune checkpoint inhibitors, time to benefit, event-free survival, overall survival, absolute risk reduction, Kaplan-Meier reconstruction, Weibull survival modeling, PD-L1, randomized controlled trials, BMC Cancer