Capturing durable benefits: interpreting six-year treatment-free survival and restricted mean survival time in CheckMate 227

Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of metastatic non-small cell lung cancer (NSCLC), with some patients remaining stable for prolonged periods after discontinuing therapy and maintaining control of their disease without further treatment. In this context, treatment-free survival (TFS) has been proposed as a novel endpoint, reflecting the duration patients spend in a treatment-free state following discontinuation of their initial therapy (1-6).

CheckMate 227 was a randomized, open-label, phase 3 trial evaluating nivolumab plus ipilimumab (nivo + ipi), nivolumab monotherapy, or chemotherapy alone as the first-line therapy for metastatic NSCLC (7,8). In Part 1 of the study, nivo + ipi demonstrated overall survival (OS) benefit over chemotherapy, irrespective of tumor programmed death-ligand 1 (PD-L1) expression. Similarly, a previous analysis involving four years of follow-up also showed that nivo + ipi prolonged TFS compared with chemotherapy (9).

The current study by Peters et al. presents an updated analysis involving more than six years of follow-up and provides a unique opportunity to assess the long-term implications of dual immunotherapy (10). The updated analysis of CheckMate 227 Part 1 provides compelling evidence of the long-term value of dual immunotherapy. At six years, the mean TFS was 12.2 months for nivo + ipi compared with 5.0 months for chemotherapy, with 17% versus 7% of the six-year period spent in TFS. These findings demonstrate that ICIs can deliver durable post-treatment benefits extending beyond the active treatment period—a phenomenon long recognized in retrospective studies and meta-analyses but rarely demonstrated so robustly in the randomized trials (11-14). Notably, the TFS benefit was observed irrespective of PD-L1 expression, thus underscoring the broad applicability of dual immunotherapy across biomarker-defined subgroups.

The study also emphasized the quality of the treatment-free period by evaluating TFS without grade ≥3 treatment-related adverse events (TRAEs). In this analysis, patients in the nivo + ipi arm experienced 11.6 months without treatment or severe toxicities compared with 4.8 months in the chemotherapy arm, underscoring the fact that extended time off treatment was not offset by a long-term toxicity burden. Moreover, the consistency of TFS findings with the previously reported OS benefit (20% versus 11% survival at six years) reinforced the validity of TFS as a clinically meaningful endpoint. In addition, the design of CheckMate 227 Part 1 had several methodological advantages for evaluating TFS. As a randomized phase 3 trial with an extensive long-term follow-up period of at least 73.5 months, it provided robust evidence that extended beyond survival to capture the post-treatment period and delayed TRAEs. Furthermore, the prespecified two-year maximum duration of immunotherapy created an ideal framework for assessing TFS by clearly distinguishing between the on-treatment and post-treatment intervals. This design feature, in turn, enabled a comprehensive assessment of patient outcomes following planned treatment discontinuation, thereby providing valuable insights into the durability of the benefits conferred by immunotherapy. Another notable strength was the use of restricted mean survival time (RMST). Unlike conventional survival analyses, RMST does not rely on the assumption of proportional hazards, making it particularly well-suited for immunotherapy trials where hazard ratios may vary over time due to delayed and sustained treatment effects (15,16). Quantifying TFS with RMST also provides a clinically intuitive interpretation, enabling direct comparison of the average time patients spend in different health states (17,18). This approach effectively captures the characteristic pattern of immunotherapy response, where curve separation may be delayed but benefits can be sustained over an extended period.

In the broader context of treatment duration, these results highlighted the potential of immunotherapy to provide benefits extending beyond the period of active treatment. The CheckMate 153 trial in the U.S. and Canada randomized patients who had tolerated one year of nivolumab to either a treatment continuation or treatment discontinuation arm (19). While progression-free survival (PFS) (PFS population: 24.7 versus 9.4 months) and OS (PFS population: not reached versus 32.5 months) favored the continuation arm, the study was designed to address safety rather than efficacy and was limited by imbalances in baseline characteristics that disadvantaged the discontinuation arm. Notably, retreatment with nivolumab after progression achieved disease control in some patients, echoing previous findings of NSCLC retreated with pembrolizumab and of melanoma retreated with ICIs (20-22). In this context, the findings of CheckMate 227 related to TFS demonstrated that treatment discontinuation did not inevitably lead to an immediate loss of benefits and that prolonged treatment-free periods were achievable in a substantial proportion of the patients.

Despite these strengths, several limitations warrant careful consideration. The open-label design of CheckMate 227 introduced potential bias in the assessment of TFS, as the decisions regarding subsequent therapy initiation relied on physician and patient discretion rather than protocol-defined criteria. This subjective element may have influenced the duration of the treatment-free period differently between treatment arms. Additionally, the definition of TFS focused exclusively on systemic therapy and did not account for local treatments such as radiotherapy or surgery. However, these limitations applied equally across treatment arms, thus preserving between-arm consistency.

Beyond trial design considerations, another interpretive caveat concerns differences in toxicity profiles between chemotherapy and immunotherapy. In the analysis of TFS by Peters et al., an important issue—the differences in the types and qualities of TRAEs between chemotherapy and immunotherapy—was not accounted for. Chemotherapy-related adverse events, such as myelosuppression, nausea, vomiting, appetite loss, and alopecia, typically manifest during active treatment and generally resolve when the treatment is interrupted or discontinued (23). In contrast, immune-related adverse events (irAEs) can present with delayed onset, even after treatment completion, and in some cases may become chronic or permanent. Endocrine irAEs requiring lifelong hormone replacement therapy exemplify this pattern (24). While severe irAEs seldom occur, their management can be challenging, and some cases may become chronic (25). Nevertheless, chemotherapy is not exempt from persistent toxicities; taxane-induced peripheral neuropathy may require considerable time to resolve, and pemetrexed-associated renal dysfunction can be irreversible (26,27). Accordingly, the real-world impact of chronic or permanent adverse events arising from immunotherapy or chemotherapy during treatment-free periods should also be considered.

RMST is informative yet not flawless, and its use in this study constitutes an inherent methodological limitation. While RMST may provide a clinically meaningful interpretation, the results depend on the chosen time horizon, and the selection of this parameter can be somewhat arbitrary (28,29). Furthermore, RMST has not yet gained widespread adoption in clinical practice, potentially limiting the immediate applicability of these findings to clinical decision-making. Although 13% of patients in the immunotherapy arm were treatment-free at six years compared with 2% in the chemotherapy arm, the absence of quality-of-life assessments and patient-reported outcome measures precludes determining whether this longer interval translated into better patient well-being.

TFS is a useful, intuitive measure of time off systemic therapy, but it does not directly capture quality of life (QOL) or patient-reported outcomes (PROs) and may underrepresent the burden of persistent and chronic irAEs. Moreover, corticosteroid use, clinic visits, procedures, and supportive care during TFS are not easily quantified. Consequently, a longer TFS does not necessarily imply greater patient benefit; interpretation should be anchored in clinical context and complemented by other endpoints. Future studies should pair TFS with QOL and PROs and with metrics of irAE burden.

Furthermore, some patients discontinued immunotherapy due to irAEs yet maintained disease control, but whether this represented a net clinical benefit or survival despite toxicity remains unclear. A more detailed evaluation of this subgroup is therefore needed. The definition of TFS itself introduced certain constraints. The reasons for treatment discontinuation—whether due to toxicity, the patient’s preference, or the physician’s decision—were not systematically analyzed. This heterogeneity, as highlighted in previous studies, may limit the objectivity and reproducibility of TFS as an endpoint. Finally, while CheckMate 227 provides evidence supporting treatment discontinuation as a viable option, questions about the optimal duration of therapy remain unanswered. The CheckMate 153 trial suggested potential disadvantages of treatment interruption, but its design was not suited to establish non-inferiority in survival. To provide clearer evidence, the ongoing JCOG1701 trial in Japan is randomizing patients with unresectable advanced or recurrent NSCLC who remain progression-free after at least one year of PD-1/PD-L1 therapy to either a treatment continuation arm or a treatment discontinuation plus observation arm (30). Unlike CheckMate 153, JCOG1701 is designed to rigorously evaluate whether a discontinuation strategy is non-inferior to continuous therapy, using OS as its primary endpoint. This trial is also expected to answer another critical question: whether ICI retreatment after progression is effective, as patients in the discontinuation arm may receive an ICI rechallenge upon disease progression. This is being formally assessed through secondary endpoints, including the response rate and PFS following retreatment. The secondary endpoints also include TFS, which will help determine whether a sustained benefit after immunotherapy discontinuation, such as that observed in CheckMate 227, can be reliably achieved by this planned discontinuation strategy.

To utilize TFS in future clinical decision-making, it can serve as vital information for patients, capturing a long-term benefit of immunotherapy that conventional endpoints like OS and PFS may miss. Particularly as OS improves and the number of elderly lung cancer patients increases, the value of TFS becomes increasingly important. It specifically quantifies the duration of time after treatment cessation when disease remains controlled without the need for subsequent therapy. For instance, the finding from CheckMate 227 means TFS was significantly longer for the nivo + ipi arm than for chemotherapy, suggesting that patients may expect a longer period free from the burdens of treatment. This prolonged period off-treatment potentially alleviates the “time toxicity” (the burden of time spent on treatment) and “financial toxicity” associated with active therapy. This holds significant economic relevance, not only for the patient but also from a health-resource utilization perspective, reducing the substantial costs of prolonged systemic therapy. This could become a valuable component of shared decision-making between physicians and patients, particularly when considering treatment discontinuation after a fixed period or when explaining long-term potential benefits at the initiation of therapy.

However, for TFS to be widely accepted as an established endpoint alongside OS and PFS, several conditions must be met. First, the standardization of the TFS definition—including when to start counting and what defines an event—and its consistent application across clinical trials are essential. Second, it is crucial to demonstrate that the patient’s QOL and PROs are well-maintained during the treatment-free period. It must be shown that this is not merely a longer duration off treatment, but a longer duration in a good health state. Third, further research must establish that an improvement in TFS strongly correlates with established primary endpoints, such as OS thereby demonstrating its validity as a surrogate endpoint. Only after such scientific evidence is accumulated and consensus is formed with regulatory authorities will TFS become established as a robust endpoint for future clinical trials.

In summary, the six-year analysis of CheckMate 227 Part 1 underscores the unique potential of dual immunotherapy to deliver prolonged TFS, often without severe TRAEs, and across PD-L1 subgroups. By complementing OS and PFS with TFS, this study provides a more comprehensive picture of the benefits of immunotherapy. Nevertheless, uncertainties about the optimal duration of therapy, the consequences of irAEs, and the subjective value of time spent off treatment remain. Ongoing prospective studies are expected to clarify these issues and guide the integration of discontinuation strategies into clinical practice. Overall, CheckMate 227 offers compelling evidence of the durable, post-treatment benefits achievable with dual immunotherapy in metastatic NSCLC.

Acknowledgments

We thank Mr. James R. Valera for his assistance with editing this manuscript.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Journal of Thoracic Disease. The article has undergone external peer review.

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2095/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2095/coif). T.K. has received personal fees from Chugai Pharmaceutical and AstraZeneca, outside the submitted work. S.N. has received personal fees from Chugai Pharmaceutical, Genmab, and AbbVie, outside the submitted work. T.H. has received personal fees from Chugai Pharmaceutical, Takeda, and Daiichi-Sankyo, outside the submitted work, and is currently an employee of GlaxoSmithKline. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Regan MM, Werner L, Rao S, et al. Treatment-Free Survival: A Novel Outcome Measure of the Effects of Immune Checkpoint Inhibition-A Pooled Analysis of Patients With Advanced Melanoma. J Clin Oncol 2019;37:3350-8. [Crossref] [PubMed]
2. Regan MM, Mantia CM, Werner L, et al. Treatment-free survival over extended follow-up of patients with advanced melanoma treated with immune checkpoint inhibitors in CheckMate 067. J Immunother Cancer 2021;9:e003743. [Crossref] [PubMed]
3. Gupta M, Stukalin I, Meyers D, et al. Treatment-Free Survival After Nivolumab vs Pembrolizumab vs Nivolumab-Ipilimumab for Advanced Melanoma. JAMA Netw Open 2023;6:e2319607. [Crossref] [PubMed]
4. Michielin O, Lalani AK, Robert C, et al. Defining unique clinical hallmarks for immune checkpoint inhibitor-based therapies. J Immunother Cancer 2022;10:e003024. [Crossref] [PubMed]
5. Regan MM, Ascierto PA MD, Lipson EJ, et al. Analysis of treatment-free survival of patients with advanced melanoma receiving nivolumab as monotherapy or in combination with relatlimab in RELATIVITY-047. J Immunother Cancer 2025;13:e012747. [Crossref] [PubMed]
6. Gupta M, Wells C, Regan MM, et al. Treatment-free Survival After First-line Therapies for Metastatic Renal Cell Carcinoma: An International Metastatic Renal Cell Carcinoma Database Consortium Analysis. Eur Urol Oncol 2025;8:171-8. [Crossref] [PubMed]
7. Hellmann MD, Paz-Ares L, Bernabe Caro R, et al. Nivolumab plus Ipilimumab in Advanced Non-Small-Cell Lung Cancer. N Engl J Med 2019;381:2020-31. [Crossref] [PubMed]
8. Brahmer JR, Lee JS, Ciuleanu TE, et al. Five-Year Survival Outcomes With Nivolumab Plus Ipilimumab Versus Chemotherapy as First-Line Treatment for Metastatic Non-Small-Cell Lung Cancer in CheckMate 227. J Clin Oncol 2023;41:1200-12. [Crossref] [PubMed]
9. Peters S, Penrod J, Li J, et al. 10P Treatment-free survival (TFS) in metastatic non-small cell lung cancer (mNSCLC) patients (pts) treated with 1L nivolumab plus ipilimumab (NIVO+IPI) or platinum doublet chemotherapy (PDC) in CheckMate (CM) 227. Ann Oncol 2022;33:S32-3.
10. Peters S, Regan MM, Paz-Ares LG, et al. Treatment-Free Survival Over 6 Years of Follow-up in Patients With Metastatic NSCLC Treated With First-Line Nivolumab Plus Ipilimumab Versus Chemotherapy in CheckMate 227 Part 1. J Thorac Oncol 2025;20:1505-16. [Crossref] [PubMed]
11. Pala L, Pagan E, Sala I, et al. Outcomes of patients with advanced solid tumors who discontinued immune-checkpoint inhibitors: a systematic review and meta-analysis. EClinicalMedicine 2024;73:102681. [Crossref] [PubMed]
12. Hu Y, Liu S, Wang L, et al. Treatment-free survival after discontinuation of immune checkpoint inhibitors in mNSCLC: a systematic review and meta-analysis. Front Immunol 2023;14:1202822. [Crossref] [PubMed]
13. Lallas K, Chatziioannou E, Durak D, et al. Discontinuation of immune checkpoint inhibitors for reasons other than disease progression and the impact on relapse and survival of advanced melanoma patients. A systematic review and meta-analysis. Front Immunol 2025;16:1524945. [Crossref] [PubMed]
14. Sun L, Bleiberg B, Hwang WT, et al. Association Between Duration of Immunotherapy and Overall Survival in Advanced Non-Small Cell Lung Cancer. JAMA Oncol 2023;9:1075-82. [Crossref] [PubMed]
15. Das A, Lin TA, Lin C, et al. Assessment of Median and Mean Survival Time in Cancer Clinical Trials. JAMA Netw Open 2023;6:e236498. [Crossref] [PubMed]
16. Uno H, Claggett B, Tian L, et al. Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. J Clin Oncol 2014;32:2380-5. [Crossref] [PubMed]
17. Perego C, Sbolli M, Specchia C, et al. Utility of Restricted Mean Survival Time Analysis for Heart Failure Clinical Trial Evaluation and Interpretation. JACC Heart Fail 2020;8:973-83. [Crossref] [PubMed]
18. Hettinger GCN, Long Q, Parikh RB. Real-World Effectiveness of Second-Line Therapies in Advanced Non-Small Cell Lung Cancer: Insights From Propensity-Weighted Comparative Analyses of Longitudinal EHR Data. Clin Lung Cancer 2025;S1525-7304(25)00236-0.
19. Waterhouse DM, Garon EB, Chandler J, et al. Continuous Versus 1-Year Fixed-Duration Nivolumab in Previously Treated Advanced Non-Small-Cell Lung Cancer: CheckMate 153. J Clin Oncol 2020;38:3863-73. [Crossref] [PubMed]
20. Herbst RS, Garon EB, Kim DW, et al. Long-Term Outcomes and Retreatment Among Patients With Previously Treated, Programmed Death-Ligand 1‒Positive, Advanced Non‒Small-Cell Lung Cancer in the KEYNOTE-010 Study. J Clin Oncol 2020;38:1580-90. [Crossref] [PubMed]
21. Robert C, Ribas A, Hamid O, et al. Durable Complete Response After Discontinuation of Pembrolizumab in Patients With Metastatic Melanoma. J Clin Oncol 2018;36:1668-74. [Crossref] [PubMed]
22. Rossi S, Masini S, Finocchiaro G, et al. Retreatment with Immune Checkpoint Inhibitors in the New Scenario of Immunotherapy in Non-Small Cell Lung Cancer. Cancers (Basel) 2024;16:1683. [Crossref] [PubMed]
23. Kuderer NM, Desai A, Lustberg MB, et al. Mitigating acute chemotherapy-associated adverse events in patients with cancer. Nat Rev Clin Oncol 2022;19:681-97. [Crossref] [PubMed]
24. Johnson DB, Nebhan CA, Moslehi JJ, et al. Immune-checkpoint inhibitors: long-term implications of toxicity. Nat Rev Clin Oncol 2022;19:254-67. [Crossref] [PubMed]
25. Barron CC, Stefanova I, Cha Y, et al. Chronic immune-related adverse events in patients with cancer receiving immune checkpoint inhibitors: a systematic review. J Immunother Cancer 2023;11:e006500. [Crossref] [PubMed]
26. Staff NP, Grisold A, Grisold W, et al. Chemotherapy-induced peripheral neuropathy: A current review. Ann Neurol 2017;81:772-81. [Crossref] [PubMed]
27. Visser S, Huisbrink J, van 't Veer NE, et al. Renal impairment during pemetrexed maintenance in patients with advanced nonsmall cell lung cancer: a cohort study. Eur Respir J 2018;52:1800884. [Crossref] [PubMed]
28. Royston P, Parmar MK. Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome. BMC Med Res Methodol 2013;13:152. [Crossref] [PubMed]
29. Tian L, Jin H, Uno H, et al. On the empirical choice of the time window for restricted mean survival time. Biometrics 2020;76:1157-66. [Crossref] [PubMed]
30. Nomura S, Goto Y, Mizutani T, et al. A randomized phase III study comparing continuation and discontinuation of PD-1 pathway inhibitors for patients with advanced non-small-cell lung cancer (JCOG1701, SAVE study). Jpn J Clin Oncol 2020;50:821-5. [Crossref] [PubMed]