Abstract
PURPOSE
The optimal second-line strategy for patients with extensive-stage small cell lung cancer (ES-SCLC) developing brain-only progression (BOP) after first-line therapy remains undefined. We aimed to evaluate the efficacy of continuing the original systemic therapy versus switching strategies.
METHODS
This multicenter cohort study screened 889 patients with ES-SCLC. A total of 203 patients developing BOP were assigned to 3 second-line strategies: continuation of original systemic therapy plus brain radiotherapy (OTP + BRT), substitution therapy plus BRT (ST + BRT), or substitution therapy alone (ST). Inverse probability of treatment weighting was used to balance baseline characteristics. The primary end point was overall survival from second-line initiation (OS2).
RESULTS
In the inverse probability of treatment weighting–weighted analysis of 203 BOP patients, OTP + BRT demonstrated significantly superior median OS2 (14.7 months) compared with ST (10.2 months; hazard ratio [HR], 1.68; P = .028) and ST + BRT (9.8 months; HR, 1.67; P = .023). OTP + BRT also yielded improved median second-line progression-free survival (PFS) (8.0 months) versus ST (4.0 months; P = .024). Multivariable analysis confirmed OTP + BRT as an independent prognostic factor for improved survival. The benefit was most pronounced in patients with prior immunotherapy and longer initial PFS (≥7.5 months). No significant survival differences were observed among radiotherapy modalities (whole-brain radiotherapy v stereotactic radiosurgery).
CONCLUSION
For ES-SCLC patients with BOP, continuing the original systemic regimen plus BRT yields superior survival compared with switching systemic therapy. This supports a site-of-progression–directed strategy, effectively controlling the CNS sanctuary while maintaining an effective systemic backbone.
INTRODUCTION
Small cell lung cancer (SCLC) is a highly aggressive neuroendocrine malignancy accounting for approximately 15% of all lung cancers worldwide and is characterized by rapid proliferation, early widespread dissemination, and a dismal prognosis.1 Extensive-stage SCLC (ES-SCLC) constitutes the majority of diagnoses and typically presents with metastatic disease; under traditional therapy, the median overall survival (OS) is approximately 7-12 months.2 In recent years, incorporation of immune checkpoint inhibitors (ICIs; anti–PD-1/PD-L1 antibodies) into first-line care has reshaped the ES-SCLC treatment landscape: pivotal trials IMpower133 and CASPIAN showed that adding atezolizumab or durvalumab to platinum-etoposide significantly improves OS and progression-free survival (PFS) over chemotherapy alone, establishing a new standard of care.3,4 This progress is grounded in the immunologic framework whereby blocking inhibitory pathways enhances T-cell–mediated cytotoxicity against tumor cells.5 Nevertheless, despite initial responses in some patients, most ultimately progress. Moreover, existing evidence suggests that first-line chemoimmunotherapy does not clearly reduce subsequent intracranial progression risk.6,7 This underscores the urgent need for effective second-line strategies to address resistance and metastatic complications. Brain metastasis is a central challenge in SCLC, affecting up to 50% of patients over the course of the disease and contributing substantially to morbidity and mortality.8 This propensity reflects the disease's intrinsic aggressiveness and the limited CNS penetration of some drugs across the blood-brain barrier (BBB).9 Historically, brain metastases have been managed with whole-brain radiotherapy (WBRT) or stereotactic radiosurgery/radiotherapy (SRS/SRT), and prophylactic cranial irradiation (PCI) has been used in select populations to reduce intracranial failure.10 In the immunotherapy era, however, the survival benefit of PCI in ES-SCLC remains controversial, and many patients still develop brain metastases after first-line therapy.11 The efficacy of immunotherapy against CNS disease is further complicated by the unique intracranial immune microenvironment, while radiotherapy may potentiate immunotherapy via immunogenic cell death and antigen release.12,13 Preclinical data support radio-immunologic synergy, yet clinical evidence in SCLC remains limited.14
CONTEXT
Key Objective
To elucidate the optimal second-line strategy for patients with extensive-stage small cell lung cancer experiencing brain-only progression (BOP) after first-line therapy, specifically comparing continued original systemic therapy plus brain radiotherapy (BRT) against switching systemic regimens.
Knowledge Generated
This multicenter cohort study demonstrates that for patients who experience BOP after first-line chemotherapy, continuing original systemic therapy in combination with BRT provided superior survival outcomes compared with switching systemic therapy, with the most pronounced benefit observed in those who received prior immunotherapy and maintained longer initial response.
Relevance (C. Chung)
The results of this study suggest that the pattern of recurrence may influence the optimal salvage therapy, supporting continuation of the original systemic therapy and adding radiation therapy to the brain for BOP.*
*Relevance section written by JCO Associate Editor Caroline Chung, MD, MSc, FRCPC.
Second-line research in ES-SCLC after progression is fragmented, particularly regarding how the site of progression (brain-only, brain plus extracranial, or extracranial-only) influences therapeutic benefit. For extracranial progression, continuing immunotherapy or switching to alternative systemic regimens (eg, topotecan or amrubicin) may extend survival, but head-to-head data are sparse and benefits are modest with nontrivial toxicity.15,16 For brain-only progression (BOP), there is no consensus on whether to continue the original systemic therapy (OTP), substitution therapy (switching systemic therapy [ST]), or integrate brain radiotherapy (BRT). Given uncertainties around immunotherapy's pharmacologic penetration and activity in the CNS and the still-emerging evidence for radio-immunotherapy synergy in SCLC,14,17 formulating site-directed, progression pattern–informed second-line strategies is especially important.
On this basis, we analyzed a real-world cohort of 889 ES-SCLC cases without baseline brain metastases across three hospitals in China to evaluate intracranial and extracranial progression and subsequent management after first-line chemoimmunotherapy or chemotherapy alone, focusing on how site-specific second-line strategies (particularly OTP + BRT v systemic switch alone) affect outcomes—aiming to provide evidence for individualized management, inform clinical practice, and guide development of new therapies.
METHODS
Ethics Approval and Consent to Participate
This study was approved by the institutional review boards of all participating centers (Shandong Cancer Hospital, Approval No. SDTHEC202512002). The requirement for written informed consent was waived due to the retrospective design. All procedures adhered to the Declaration of Helsinki.
Study Design and Patients
This multicenter retrospective cohort study was conducted at three tertiary cancer centers in China (SDCH, SHCH, and TJCH). We screened 889 ES-SCLC patients without baseline brain metastases who received standard first-line platinum-etoposide chemotherapy (with or without immune checkpoint inhibitors). The final analytic cohort focused specifically on the 203 patients who developed BOP after first-line therapy. Detailed inclusion and exclusion criteria, along with the patient selection flowchart, are presented in Figure 1 and the Data Supplement (online only).
FIG 1.
A flowchart illustrating the patient selection process for the study on ES-SCLC. ES-SCLC, extensive-stage small-cell lung cancer.
Definition and Ascertainment of Brain-Only Progression
BOP was strictly defined as radiographically confirmed intracranial progression (per RANO-BM criteria) in the presence of stable or responsive extracranial disease (per RECIST v1.1). To strictly exclude concurrent systemic failure, paired contrast-enhanced brain magnetic resonance imaging (MRI) and extracranial computed tomography (CT) scans were required to be performed within a <1-week interval. Patients with leptomeningeal disease, any evidence of concurrent extracranial progression (including oligoprogression), or prior PCI were excluded.
Treatment Protocol
Patients with BOP were categorized into 3 second-line strategies based on physician discretion:
OTP + BRT: Continuation of the original systemic therapy (maintaining the immune checkpoint inhibitor or chemotherapy backbone) combined with BRT.
ST + BRT: Switching to a new systemic therapy (substitution therapy) combined with BRT.
ST: Switching to a new systemic therapy alone.
For patients in the OTP + BRT and ST + BRT groups, detailed radiotherapy modalities (WBRT, whole-brain radiotherapy with simultaneous integrated boost [WBRT-SIB], or SRS/SRT) and dosimetry data are provided in the Data Supplement.
Clinical Response Assessment
Tumor responses were assessed using paired brain MRI and whole-body CT. Extracranial status was evaluated according to RECIST v1.1, while intracranial response was assessed using RANO-BM criteria. Surveillance imaging was typically performed every 8-12 weeks or upon clinical deterioration. Imaging intervals were reviewed to ensure comparable surveillance intensity across treatment groups.
End Points
To eliminate immortal-time bias, the primary end point was overall survival 2 (OS2), defined as the time from the initiation of the second-line treatment strategy (specifically, the start of BRT for the OTP + BRT group, or the start of the substitution systemic therapy for the ST and ST + BRT groups) to death from any cause. The key secondary end point was second-line progression-free survival (PFS2). OS from initial diagnosis was analyzed as a secondary outcome.
Statistical Analysis
To mitigate selection bias, we used inverse probability of treatment weighting (IPTW) based on covariate balancing propensity scores (CBPS). The propensity model incorporated prespecified covariates, including study center, calendar year, first-line regimen, iPFS (modeled continuously), and metastatic burden (details in the Data Supplement). Covariate balance was confirmed by standardized mean differences (SMD <0.10, Data Supplement, Fig S1).
Survival outcomes were estimated using weighted Kaplan-Meier methods and compared via weighted log-rank tests. Treatment effects were quantified using weighted Cox proportional hazards models. To account for center-level heterogeneity, sensitivity analyses were performed using shared frailty (mixed-effects) Cox models with study center as a random intercept (Data Supplement, Table S3). P values for confirmatory analyses were adjusted using the Holm-Bonferroni method (Data Supplement, Table S11). Missing data were not imputed as the data set was complete (Data Supplement, Table S4). Analyses were performed using R (version 4.2.3; R Foundation for Statistical Computing) and GraphPad Prism (version 8.0.2; GraphPad Software).
RESULTS
Patient Characteristics
The study follow-up continued until March 10, 2025. The full cohort comprised 889 patients with ES-SCLC who had progressed after standard first-line platinum-etoposide chemotherapy (with or without PD-L1/PD-1 inhibitors). Baseline characteristics of the full cohort are shown in the Data Supplement (Table S5). The median age was 64 years, and 88.98% had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0-1. Patients were categorized by first-line immunotherapy usage: 443 (49.83%) received chemoimmunotherapy (immunotherapyi [IO] + ChT), and 446 (50.17%) received chemotherapy alone (ChT). The overall objective response rate to first-line therapy was 76.38%.
Incidence of BMs and First-Line Progression Sites
The median follow-up duration for the entire cohort was 33.1 months. Consistent with pivotal trials, patients in the IO + ChT group achieved significantly longer median PFS and OS compared with the ChT group (Data Supplement, Fig S4A and S4B). However, pattern-of-failure analysis revealed that the addition of immunotherapy did not reduce the cumulative incidence of subsequent brain metastases (33.0% in IO + ChT v 33.4% in ChT; P = .239; Data Supplement, Fig S4C). The brain remained the second most frequent site of progression (295 patients, 33.2%), after locoregional thoracic recurrence (36.8%; Fig 2). From this population, 203 patients met the strict criteria for BOP—defined as intracranial progression with maintained extracranial control—and constituted the final analytic cohort.
FIG 2.
Baseline metastasis sites and first-line progression sites in the study cohort (n = 889). Metastatic and progression sites were not mutually exclusive. BOP, brain-only progression.
Baseline Characteristics of Patients With BOP
The 203 BOP patients were categorized into 3 second-line strategies: OTP + BRT (n = 71), ST + BRT (n = 64), and ST (n = 68). Before weighting, significant imbalances were observed: patients in the OTP + BRT group had a lower frequency of baseline liver metastases and prior immunotherapy exposure but longer initial PFS (iPFS) compared with the substitution groups (Table 1). After applying CBPS-IPTW, baseline characteristics were substantially balanced among the three groups. The weighted populations showed comparable distributions of age, sex, ECOG PS, smoking history, baseline liver and bone metastases, prior immunotherapy exposure, intracranial tumor burden, treatment center, calendar year at BOP, iPFS, and first-line chemotherapy exposure. Absolute SMDs were reduced to <0.10 for most covariates and <0.20 for all covariates, indicating adequate covariate balance for subsequent weighted outcome analyses (Data Supplement, Table S6).
TABLE 1.
Baseline Characteristics of Patients With Intracranial Progression
| Covariate | OTP + BRT | ST | ST + BRT | P | SMD |
|---|---|---|---|---|---|
| Total No. | 71 | 68 | 64 | ||
| Sex | |||||
| Male | 59.0 (83.1) | 57.0 (83.8) | 55.0 (85.9) | .897 | 0.078 |
| Female | 12.0 (16.9) | 11.0 (16.2) | 9.0 (14.1) | ||
| Age, years | |||||
| <65 | 36.0 (50.7) | 27.0 (39.7) | 38.0 (59.4) | .079 | 0.399 |
| ≥65 | 35.0 (49.3) | 41.0 (60.3) | 26.0 (40.6) | ||
| Smoking status | |||||
| No | 23.0 (32.4) | 24.0 (35.3) | 20.0 (31.2) | .878 | 0.086 |
| Yes | 48.0 (67.6) | 44.0 (64.7) | 44.0 (68.8) | ||
| ECOG | |||||
| ECOG: 0-1 | 63.0 (88.7) | 66.0 (97.1) | 60.0 (93.8) | .148 | 0.333 |
| ECOG: 2 | 8.0 (11.3) | 2.0 (2.9) | 4.0 (6.2) | ||
| Baseline liver metastases | |||||
| No | 61.0 (85.9) | 47.0 (69.1) | 44.0 (68.8) | .031 | 0.401 |
| Yes | 10.0 (14.1) | 21.0 (30.9) | 20.0 (31.2) | ||
| Baseline bone metastases | |||||
| No | 52.0 (73.2) | 49.0 (72.1) | 47.0 (73.4) | .981 | 0.031 |
| Yes | 19.0 (26.8) | 19.0 (27.9) | 17.0 (26.6) | ||
| Best efficacy evaluation | |||||
| PR | 56.0 (78.9) | 54.0 (79.4) | 56.0 (87.5) | .249 | 0.225 |
| SD | 13.0 (18.3) | 14.0 (20.6) | 8.0 (12.5) | ||
| CR | 2.0 (2.8) | 0.0 (0.0) | 0.0 (0.0) | ||
| Immunotherapy | |||||
| No | 50.0 (70.4) | 35.0 (51.5) | 27.0 (42.2) | .004 | 0.584 |
| Yes | 21.0 (29.6) | 33.0 (48.5) | 37.0 (57.8) | ||
| Number of brain metastases | |||||
| 1 | 22.0 (31.0) | 17.0 (25.0) | 17.0 (26.6) | .935 | 0.134 |
| 2-3 | 14.0 (19.7) | 13.0 (19.1) | 12.0 (18.8) | ||
| >3 | 35.0 (49.3) | 38.0 (55.9) | 35.0 (54.7) | ||
| Center | |||||
| 1: SDCH | 36.0 (50.7) | 47.0 (69.1) | 27.0 (42.2) | .000 | 0.555 |
| 2: TJCH | 25.0 (35.2) | 4.0 (5.9) | 14.0 (21.9) | ||
| 3: SHCH | 10.0 (14.1) | 17.0 (25.0) | 23.0 (35.9) | ||
| iPFS: mean (SD) | 8.3 (4.8) | 10.4 (6.9) | 8.1 (4.6) | .063 | 0.408 |
| Calendar year (BOP) | |||||
| 2019 | 0.0 (0.0) | 1.0 (1.5) | 0.0 (0.0) | .216 | 0.212 |
| 2020 | 4.0 (5.6) | 2.0 (2.9) | 0.0 (0.0) | ||
| 2021 | 7.0 (9.9) | 10.0 (14.7) | 15.0 (23.4) | ||
| 2022 | 14.0 (19.7) | 10.0 (14.7) | 14.0 (21.9) | ||
| 2023 | 31.0 (43.7) | 27.0 (39.7) | 26.0 (40.6) | ||
| 2024 | 15.0 (21.1) | 18.0 (26.5) | 9.0 (14.1) | ||
| First-line EP cycles | |||||
| <4/>6 | 13.0 (18.3) | 12.0 (17.6) | 3.0 (4.7) | .040 | 0.405 |
| 4-6 | 58.0 (81.7) | 56.0 (82.4) | 61.0 (95.3) | ||
| RT modality | |||||
| WBRT-SIB | 27.0(38.0) | — | 24.0(37.5) | .994 | 0.018 |
| WBRT | 29.0(40.8) | — | 26.0(40.6) | ||
| SRS/SRT | 15.0(21.2) | — | 14.0(21.9) |
NOTE. Data are presented as No. (%) for categorical variables and mean (SD) for continuous variables. Bold values indicate statistically significant between-group differences, defined as P < .05. P-values were calculated using the chi-square test or Fisher's exact test for categorical variables, and analysis of variance test for continuous variables.
Abbreviations: BOP, brain-only progression; BRT, brain radiotherapy; CBPS, covariate balancing propensity score; CR, complete response; ECOG, Eastern Cooperative Oncology Group; EP, etoposide plus platinum; iPFS, intracranial progression-free survival; OTP, original systemic therapy; PR, partial response; SD, stable disease; SDCH, Shandong Cancer Hospital; SHCH, Shanghai Chest Hospital; SIB, simultaneous integrated boost; SMD, standardized mean difference; SRS/SRT, stereotactic radiosurgery/stereotactic radiotherapy; ST, systemic therapy; TJCH, Tianjin Cancer Hospital; WBR, whole-brain radiotherapy; WBRT-SIB, whole-brain radiotherapy with simultaneous integrated boost.
Survival Analysis of Patients With BOP
Survival outcomes differed significantly according to second-line strategy, with unweighted results mirroring the weighted findings reported here. In the rigorous IPTW-weighted analysis (Fig 3), the OTP + BRT group demonstrated the most durable disease control. For the secondary end point PFS2, the weighted median duration was 8.0 months in the OTP + BRT group, significantly longer than the 4.0 months observed in the ST group (weighted log-rank P = .024; Fig 3A), while the ST + BRT group (median 5.0 months) did not differ significantly from the OTP + BRT group (P = .179). For the primary end point OS2 calculated from second-line initiation, OTP + BRT provided superior benefit with a weighted median OS2 of 14.7 months. By contrast, both switching strategies resulted in significantly inferior survival: 10.2 months for the ST group (P = .028; Fig 3B) and 9.8 months for the ST + BRT group (P = .023; Fig 3B). Cumulative OS from diagnosis showed a similar trend, with OTP + BRT (median 26.6 months) significantly outperforming ST + BRT (20.0 months; P = .022; Fig 3C) and numerically outperforming ST (17.6 months; P = .052; Fig 3C). Notably, the survival disadvantage for both ST (adjusted P = .046) and ST + BRT (Adjusted P = .046) regarding the primary end point OS2 remained statistically significant after Holm-Bonferroni correction for multiple comparisons (Data Supplement, Table S11).
FIG 3.
Kaplan-Meier curves comparing the 3 second-line treatment strategies—continuation of OTP + BRT, ST + BRT, and ST—in the (A, C, E) unweighted and (B, D, F) weighted populations. (A and B) PFS2. (C and D) OS2 (primary end point). (E and F) OS. (B, D, F) Weighted analyses were performed using covariate balancing propensity score weighting to account for baseline imbalances between treatment groups. P values were calculated using the log-rank test for unweighted analyses and the weighted log-rank test for weighted analyses. Shaded areas represent 95% CIs. BRT, brain radiotherapy; OS, overall survival; OS2, overall survival after second-line therapy; PFS2, second-line progression-free survival.
Subgroup analysis revealed that the survival benefit of OTP + BRT was particularly pronounced in patients with prior immunotherapy exposure. In this population, weighted analysis confirmed significantly elevated mortality risks for both ST (OS2 hazard ratio [HR], 2.32; P = .038) and ST + BRT (OS2 HR, 2.42; P = .018) compared with OTP + BRT, alongside inferior PFS2 for the ST group (HR, 2.25; P = .049; Data Supplement, Fig S6). Conversely, these survival distinctions were attenuated and less evident among patients who received first-line chemotherapy alone (Data Supplement, Fig S7).
Outcomes in Short Versus Long iPFS Subgroups
Patients were stratified into short (<7.5 months) and long (≥7.5 months) intracranial PFS (iPFS) subgroups based on the median time to brain metastasis. Baseline clinical characteristics, including age, ECOG PS, and best response to first-line therapy, were balanced between these subgroups (Data Supplement, Table S7). In the short iPFS subgroup, weighted analyses showed no statistically significant differences in PFS2 or OS2 among the OTP + BRT, ST, and ST + BRT strategies (all P > .30; Data Supplement, Figs S8A and S8C). In the long iPFS subgroup, OTP + BRT resulted in significantly longer median PFS2 (11.0 vs 4.0 months; P = .046; Data Supplement, Fig S8B) and median OS2 (19.4 vs 9.4 months; P = .001; Data Supplement, Fig S8D) compared with ST.
Intracranial Efficacy and Impact of Radiotherapy Modality on Survival Outcomes
Our analysis was further enriched by a comprehensive characterization of patients' intracranial metastatic burden, encompassing the number and maximum diameter of brain metastases, alongside detailed information on radiotherapy strategies and dosage regimens, all meticulously documented in the Data Supplement. We further evaluated intracranial control and radiotherapy techniques using weighted analyses. There were no statistically significant differences in iPFS among the three groups (OTP + BRT: not reached; ST: 17.3 months; ST + BRT: 25.9 months; weighted P > .05; Data Supplement), indicating that systemic strategy switching did not independently enhance intracranial control. Furthermore, comparisons among WBRT, WBRT-SIB, and SRS/SRT modalities revealed no significant differences in OS2, PFS2, or iPFS (Data Supplement, Fig S11), implying that effective local control—regardless of the specific modality—is the critical determinant of survival outcomes.
Prognostic Factors for PFS2 and OS2
The independent prognostic value of treatment strategies and baseline variables was evaluated using IPTW-weighted multivariable Cox regression models. For PFS2, second-line treatment strategy was an independent predictor; compared with OTP + BRT, switching to ST was associated with a significantly higher risk of progression (HR, 1.69 [95% CI, 1.12 to 2.53]; P = .012), while other risk factors included ECOG PS of 2 (HR, 2.55; P = .003) and stable disease as the best response to first-line therapy (HR, 2.12; P < .001) (Data Supplement, Table S8). For the primary end point OS2, both alternative strategies were independently associated with increased mortality risk compared with OTP + BRT, with adjusted hazard ratios of 1.75 for ST (95% CI, 1.11 to 2.77; P = .016) and 1.80 for ST + BRT (95% CI, 1.16 to 2.79; P = .009); consistent with PFS2, stable disease after first-line therapy (HR, 2.33; P < .001) was a strong adverse prognostic factor, whereas longer initial PFS duration (HR, 1.07 per month; P < .001) was independently associated with improved survival (Table 2).
TABLE 2.
Weighted Univariate and Multivariate Cox Regression Analyses of Factors Influencing OS2 in Patients With Intracranial Progression
| Variables | Univariate Analysis | Multivariate Analysis | ||
|---|---|---|---|---|
| HR (95% CI) | P | HR (95% CI) | P | |
| Group | ||||
| OTP + BRT | 1.00 (reference) | |||
| ST | 1.67 (1.05 to 2.68) | .032 | 1.75 (1.11 to 2.77) | .016 |
| ST + BRT | 1.66 (1.10 to 2.53) | .017 | 1.80 (1.16 to 2.79) | .009 |
| Sex | ||||
| Male | 1.00 (reference) | |||
| Female | 0.71 (0.44 to 1.16) | .171 | ||
| Age, years | ||||
| ≥65 | 1.00 (reference) | |||
| <65 | 1.29 (0.90 to 1.85) | .174 | ||
| Smoking | ||||
| No | 1.00 (reference) | |||
| Yes | 1.14 (0.79 to 1.65) | .474 | ||
| ECOG | ||||
| 0-1 | 1.00 (reference) | |||
| 2 | 1.58 (0.97 to 2.59) | .068 | 1.73 (0.97 to 3.08) | .062 |
| Baseline liver metastases | ||||
| No | 1.00 (reference) | |||
| Yes | 1.13 (0.74 to 1.73) | .559 | ||
| Baseline bone metastases | ||||
| No | 1.00 (reference) | |||
| Yes | 1.23 (0.82 to 1.85) | .327 | ||
| Number of brain metastases | ||||
| 1 | 1.00 (reference) | |||
| 2-3 | 1.38 (0.82 to 2.31) | .223 | ||
| >3 | 1.20 (0.76 to 1.88) | .433 | ||
| Best efficacy evaluation | ||||
| PR | 1.00 (reference) | |||
| SD | 2.12 (1.41 to 3.20) | .000 | 2.33 (1.53 to 3.54) | .000 |
| CR | 0.28 (0.06 to 1.37) | .115 | 0.24 (0.02 to 2.29) | .213 |
| Immunotherapy | ||||
| No | 1.00 (reference) | |||
| Yes | 0.80 (0.55 to 1.16) | .240 | ||
| Center | ||||
| 1: SDCH | 1.00 (reference) | |||
| 2: TJCH | 0.84 (0.56 to 1.26) | .397 | ||
| 3: SHCH | 1.06 (0.67 to 1.69) | .796 | ||
| iPFS | 1.06 (1.02 to 1.11) | .006 | 1.07 (1.03 to 1.11) | .000 |
| Calendar year | ||||
| 2019 | 1.00 (reference) | |||
| 2020 | 0.36 (0.23 to 0.56) | .000 | 0.26 (0.14 to 0.51) | .000 |
| 2021 | 0.47 (0.30 to 0.73) | .001 | 0.37 (0.20 to 0.68) | .001 |
| 2022 | 0.63 (0.42 to 0.95) | .028 | 0.47 (0.30 to 0.75) | .002 |
| 2023 | 0.62 (0.45 to 0.86) | .004 | 0.52 (0.33 to 0.81) | .003 |
| 2024 | 0.35 (0.19 to 0.67) | .001 | 0.31 (0.15 to 0.64) | .002 |
| EP cycles | ||||
| <4/>6 | 1.00 (reference) | |||
| 4-6 | 0.95 (0.56 to 1.62) | .848 | ||
NOTE. Data represent the results from the propensity score–weighted Cox proportional hazards regression model. Multivariate analysis adjusted for covariates with P < .10 in the univariate analysis. Bold values in the univariate analysis indicate variables meeting the prespecified threshold for multivariable model inclusion, defined as P < .10. Bold values in the multivariable analysis indicate statistical significance, defined as P < .05.
Abbreviations: BRT, brain radiotherapy; CR, complete response; ECOG PS, Eastern Cooperative Oncology Group performance status; EP, etoposide plus platinum; HR, hazard ratio; iPFS, intracranial progression-free survival; OS2, overall survival 2; OTP, original systemic therapy; PR, partial response; SD, stable disease; SDCH, Shandong Cancer Hospital; SHCH, Shanghai Chest Hospital; ST, systemic therapy; TJCH, Tianjin Cancer Hospital.
Exploratory subgroup analyses within the weighted cohort demonstrated that the survival advantage of OTP + BRT over ST + BRT was generally consistent across key clinical profiles (overall HR, 1.65 [95% CI, 1.08 to 2.50]). Formal interaction tests revealed no statistically significant heterogeneity regarding baseline characteristics, including intracranial tumor burden (P for interaction = .323) and radiotherapy modality. Although the magnitude of benefit appeared numerically greater in patients with longer initial PFS (HR, 1.75 v 1.30; P = .070) and prior immunotherapy exposure (HR, 2.40 v 1.33; P = .183), these differences did not reach statistical significance. These findings support the robustness of the primary results across the study population (Data Supplement, Fig S5).
DISCUSSION
In this three-center, real-world cohort of 889 patients with ES-SCLC without baseline brain metastases, first-line chemoimmunotherapy improved OS but did not reduce the subsequent risk of brain metastasis. Among patients who developed BOP, continuing the original systemic therapy combined with BRT (OTP + BRT) significantly prolonged second-line PFS2 and OS2 compared with substitution therapy alone (ST). The greatest benefit was observed in patients whose first-line regimen included immunotherapy (median OS 38.28 months in the immunotherapy subgroup). These findings support a site-of-progression–centered second-line strategy: when extracranial disease remains controlled, clinicians should prioritize addressing the lack of CNS disease control with radiation therapy rather than switching systemic therapy.
The lack of intracranial protection with first-line chemoimmunotherapy in our cohort is consistent with the IMpower133 and CASPIAN trials, which established platinum-etoposide plus a PD-L1 inhibitor as the standard of care but did not demonstrate a clear reduction in the risk of intracranial progression.3,4 Prior pooled analyses likewise indicated that adding anti–PD-L1 improved OS without decreasing the incidence of subsequent brain metastases, underscoring that systemic survival benefit does not necessarily translate into intracranial disease control or a reduced risk of CNS progression.6 Patients with BOP appear to harbor micrometastatic foci within the CNS, which may progress independently despite systemic disease control, suggesting a limited role for ongoing hematogenous dissemination during treatment. This phenomenon suggests a sanctuary effect driven by the BBB and the intracranial immune microenvironment, which limits drug penetration and efficacy in the brain while extracranial lesions remain controlled.9,18 Within this framework, simply switching systemic therapy may not fully overcome the CNS sanctuary, particularly when the substituted agents lack meaningful CNS activity; only regimens with proven CNS penetration and activity (eg, tarlatamab) are likely to address this limitation.19 In our cohort, continuing the original systemic regimen while adding BRT (OTP + BRT) was associated with more pronounced clinical benefit. A plausible explanation is that OTP + BRT directly clears resistant or occult intracranial disease, effectively controlling this sanctuary site, while preserving the systemic disease control and immune memory achieved with the established regimen, thereby potentially limiting systemic reseeding that might result from premature switching.20,21 Radiotherapy may further synergize with checkpoint blockade through antigen release, dendritic-cell activation, and T-cell clonal expansion, producing abscopal effects and a coupled local-systemic benefit.22,23
We found that, in weighted analyses, the various second-line treatment strategies did not demonstrate statistically significant differences in intracranial PFS. Although the OTP + BRT group exhibited the longest median iPFS, this difference did not reach statistical significance when compared with either the ST or ST + BRT groups. This observation suggests that the potent local control offered by BRT might mask or diminish the direct impact of different systemic therapy strategies on intracranial lesions. When local progression is effectively suppressed, the primary role of systemic therapy may shift toward maintaining extracranial disease control and influencing long-term OS benefits, rather than directly extending intracranial PFS. This aligns with the previously discussed CNS sanctuary effect, where the unique biology and BBB of established brain metastases can limit the direct clearance efficacy of systemic therapies on intracranial lesions,9,18 positioning radiotherapy as the critical modality for local eradication.
Furthermore, our evaluation of different radiotherapy modalities (WBRT, WBRT-SIB, and SRS/SRT) revealed no significant differences in iPFS, OS2, or OS. This indicates that for SCLC patients with brain metastases, the choice of radiotherapy modality may primarily depend on individual patient characteristics, tumor burden, and toxicity management strategies, rather than fundamentally affecting OS. Modern radiotherapy techniques, such as SRS/SRT, are increasingly favored for localized brain metastases due to their precision and brain-sparing capabilities, aiming to minimize neurocognitive toxicity.10,23,24 Prior studies have also indicated that, with stringent volumetric and dose constraints, SRS/SRT can achieve survival and intracranial control comparable to cohorts with fewer lesions, even in patients with ≥5-10 metastases.25,26 Our findings therefore suggest that multiple radiotherapy modalities may offer comparable efficacy in terms of survival prolongation. This provides clinicians with greater flexibility in selecting a radiotherapy strategy, allowing for a better balance between effective local control and patient quality of life.
Therapeutic context further supports this site-specific approach. Second-line options for SCLC provide modest benefit and meaningful toxicity: topotecan, the historical standard, confers limited survival advantage and tolerability in randomized trials and systematic reviews,27,28 while lurbinectedin has activity in single-arm studies but its sequential, cross-line value remains to be fully defined.29 By restoring intracranial control while maintaining a systemically active backbone, OTP + BRT can delay entry into a more toxic and less certain second-line phase.
Heterogeneity in benefit provides practical guidance for patient selection. The advantage of OTP + BRT was concentrated in patients with longer first-line PFS (iPFS ≥7.50 months), a phenotype suggestive of relatively indolent tumor biology or heightened sensitivity to initial systemic therapy—particularly immunotherapy—with the potential for durable responses.1,20 Notably, baseline clinical characteristics were generally well balanced between the short and long iPFS groups. Thus, the inferior outcomes observed in patients with short iPFS are unlikely to be explained simply by imbalances in baseline factors or depth of initial response. Instead, early intracranial failure despite ongoing extracranial control may serve as a clinical marker of intrinsically aggressive, treatment-refractory biology, in which CNS escape occurs early and may herald imminent systemic failure.30,31 Early intracranial progression can also impair neurologic function and limit the feasibility of further systemic therapy, thereby contributing to worse OS in the short iPFS subgroup.
This retrospective study is subject to selection bias despite multivariable adjustment; treatment decisions were at physician discretion and may reflect unmeasured patient or disease factors. Although the overall sample size for this clinical scenario was substantial, statistical power was limited in some subgroup analyses (eg, OTP + BRT v ST + BRT within the chemotherapy-only cohort). Moreover, cognitive and neurocognitive function outcomes, as well as patient-reported quality-of-life measures, were not systematically collected, precluding a direct assessment of the potential neurocognitive impact of different BRT strategies. The absence of centralized radiologic review and molecular or immunologic biomarkers (eg, PD-L1 expression, tumor mutational burden) limits precision in identifying patients most likely to benefit.
In conclusion, when BOP develops after first-line therapy in ES-SCLC, OTP + BRT provides significant and clinically meaningful improvements in disease control and survival compared with immediate systemic switching, with particularly pronounced benefits in patients whose first-line regimen included immunotherapy. These findings support a pragmatic, site-specific strategy—stabilize the CNS first while maintaining a systemically effective backbone—and motivate prospective randomized trials with biomarker programs to validate and refine patient selection.
See accompanying Editorial, p. 2067
SUPPORT
Supported by the National Natural Science Foundation of China [grant number: 82473254]; and Clinical Research Pioneering Program of Shandong First Medical University & Shandong Academy of Medical Sciences (607D25022).
S.L. and J.J. contributed equally to this work.
Supplementary Materials
DATA SHARING STATEMENT
The data sets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
AUTHOR CONTRIBUTIONS
Conception and design: Shuangqing Lu, Jingyi Jia, Kai Ren, Lujun Zhao, Hui Zhu
Administrative support: Kai Ren
Provision of study materials or patients: Jingyi Jia, Kai Ren, Wenxiao Jia, Xuwei Cai
Collection and assembly of data: Shuangqing Lu, Jingyi Jia, Kai Ren, Zhuoran Sun, Ke Zhao, Xuwei Cai, Hui Zhu
Data analysis and interpretation: Shuangqing Lu, Jingyi Jia, Kai Ren, Wenxiao Jia, Hui Zhu
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Multicenter Cohort Study of Original or Substitute Systemic Therapy With or Without Brain Radiotherapy for Extensive-Stage Small Cell Lung Cancer With Brain-Only Progression After First-Line Treatment
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
No potential conflicts of interest were reported.
REFERENCES
- 1.Rudin CM, Brambilla E, Faivre-Finn C, et al. : Small-cell lung cancer. Nat Rev Dis Primers 7:3, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Liu SV, Reck M, Mansfield AS, et al. : Updated overall survival and PD-L1 subgroup analysis of patients with extensive-stage small-cell lung cancer treated with atezolizumab, carboplatin, and etoposide (IMpower133). J Clin Oncol 39:619-630, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Paz-Ares L, Dvorkin M, Chen Y, et al. : Durvalumab plus platinum–etoposide versus platinum–etoposide in first-line treatment of extensive-stage small-cell lung cancer (CASPIAN): A randomised, controlled, open-label, phase 3 trial. Lancet 394:1929-1939, 2019 [DOI] [PubMed] [Google Scholar]
- 4.Horn L, Mansfield AS, Szczęsna A, et al. : First-line atezolizumab plus chemotherapy in extensive-stage small-cell lung cancer. N Engl J Med 379:2220-2229, 2018 [DOI] [PubMed] [Google Scholar]
- 5.Chen DS, Mellman I: Oncology meets immunology: The cancer-immunity cycle. Immunity 39:1-10, 2013 [DOI] [PubMed] [Google Scholar]
- 6.Lu S, Guo X, Li Y, et al. : Antiprogrammed death ligand 1 therapy failed to reduce the risk of developing brain metastases in patients with extensive-stage small cell lung cancer: A retrospective analysis. Cancer 130:18-30, 2024 [DOI] [PubMed] [Google Scholar]
- 7.Zhou Y, Guo T, Liang F, et al. : Cumulative incidence and risk factors of brain metastases in metastatic non-small cell lung cancer without baseline brain metastasis: Pooled analysis of individualized patient data from IMpower130, IMpower131, and IMpower150. Cancer 130:2601-2610, 2024 [DOI] [PubMed] [Google Scholar]
- 8.Li H, Zhao Y, Ma T, et al. : Radiotherapy for extensive-stage small-cell lung cancer in the immunotherapy era. Front Immunol 14:1132482, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Arvanitis CD, Ferraro GB, Jain RK: The blood–brain barrier and blood–tumour barrier in brain tumours and metastases. Nat Rev Cancer 20:26-41, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gondi V, Bauman G, Bradfield L, et al. : Radiation therapy for brain metastases: An ASTRO clinical practice guideline. Pract Radiat Oncol 12:265-282, 2022 [DOI] [PubMed] [Google Scholar]
- 11.Takahashi T, Yamanaka T, Seto T, et al. : Prophylactic cranial irradiation versus observation in patients with extensive-disease small-cell lung cancer: A multicentre, randomised, open-label, phase 3 trial. Lancet Oncol 18:663-671, 2017 [DOI] [PubMed] [Google Scholar]
- 12.Lynch C, Pitroda SP, Weichselbaum RR: Radiotherapy, immunity, and immune checkpoint inhibitors. Lancet Oncol 25:e352-e362, 2024 [DOI] [PubMed] [Google Scholar]
- 13.Demaria S, Golden EB, Formenti SC: Role of local radiation therapy in cancer immunotherapy. JAMA Oncol 1:1325-1332, 2015 [DOI] [PubMed] [Google Scholar]
- 14.Pointer KB, Pitroda SP, Weichselbaum RR: Radiotherapy and immunotherapy: Open questions and future strategies. Trends Cancer 8:9-20, 2022 [DOI] [PubMed] [Google Scholar]
- 15.O’Brien MER, Ciuleanu TE, Tsekov H, et al. : Phase III trial comparing supportive care alone with supportive care with oral topotecan in patients with relapsed small-cell lung cancer. J Clin Oncol 24:5441-5447, 2006 [DOI] [PubMed] [Google Scholar]
- 16.Zhang J, Guo JX, Li P, et al. : Real-world outcomes of immune checkpoint inhibitors as second-line therapy for extensive-stage small-cell lung cancer: A multicenter retrospective analysis. Front Immunol 16:1658017, 2025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Peters S, Pujol JL, Dafni U, et al. : Consolidation nivolumab and ipilimumab versus observation in limited-disease small-cell lung cancer after chemo-radiotherapy—Results from the randomised phase II ETOP/IFCT 4-12 STIMULI trial. Ann Oncol 33:67-79, 2022 [DOI] [PubMed] [Google Scholar]
- 18.Gavrilovic IT, Posner JB: Brain metastases: Epidemiology and pathophysiology. J Neurooncol 75:5-14, 2005 [DOI] [PubMed] [Google Scholar]
- 19.Zhang B, Shah KB, Parma M, et al. : Rapid intracranial response with tarlatamab in patients with untreated brain metastases from SCLC—A real-world case series: Case report. J Thorac Oncol 20:1131-1138, 2025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.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 9:1075, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Twyman-Saint Victor C, Rech AJ, Maity A, et al. : Radiation and dual checkpoint blockade activate non-redundant immune mechanisms in cancer. Nature 520:373-377, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Weichselbaum RR, Liang H, Deng L, et al. : Radiotherapy and immunotherapy: A beneficial liaison? Nat Rev Clin Oncol 14:365-379, 2017 [DOI] [PubMed] [Google Scholar]
- 23.Brown PD, Pugh S, Laack NN, et al. : Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: A randomized, double-blind, placebo-controlled trial. Neuro Oncol 15:1429-1437, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gong X, Li X, Jiang T, et al. : Combined radiotherapy and anti–PD-L1 antibody synergistically enhances antitumor effect in non–small cell lung cancer. J Thorac Oncol 12:1085-1097, 2017 [DOI] [PubMed] [Google Scholar]
- 25.Shuto T, Akabane A, Yamamoto M, et al. : Multiinstitutional prospective observational study of stereotactic radiosurgery for patients with multiple brain metastases from non–small cell lung cancer (JLGK0901 study–NSCLC). J Neurosurg 129:86-94, 2018. (suppl 1) [DOI] [PubMed] [Google Scholar]
- 26.Vogelbaum MA, Brown PD, Messersmith H, et al. : Treatment for brain metastases: ASCO-SNO-ASTRO guideline. J Clin Oncol 40:492-516, 2022 [DOI] [PubMed] [Google Scholar]
- 27.Von Pawel J, Schiller JH, Shepherd FA, et al. : Topotecan versus cyclophosphamide, doxorubicin, and vincristine for the treatment of recurrent small-cell lung cancer. J Clin Oncol 17:658, 1999 [DOI] [PubMed] [Google Scholar]
- 28.Qin K, Gay CM, Byers LA, et al. : The current and emerging immunotherapy paradigm in small-cell lung cancer. Nat Cancer 6:954-966, 2025 [DOI] [PubMed] [Google Scholar]
- 29.Cheng Y, Wu C, Wu L, et al. : A pivotal bridging study of lurbinectedin as second-line therapy in chinese patients with small cell lung cancer. Sci Rep 14:3598, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Petty WJ, Paz-Ares L: Emerging strategies for the treatment of small cell lung cancer: A review. JAMA Oncol 9:419-429, 2023 [DOI] [PubMed] [Google Scholar]
- 31.Gay CM, Stewart CA, Park EM, et al. : Patterns of transcription factor programs and immune pathway activation define four major subtypes of SCLC with distinct therapeutic vulnerabilities. Cancer Cell 39:346-360.e7, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data sets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.