Summary
Concurrent chemoradiotherapy followed by consolidation immunotherapy is standard of care (SOC) for unresectable locally advanced non-small cell lung cancer. This retrospective study enrolled 168 patients to compare survival and safety outcomes of the addition of induction therapy compared with the cCRT+ICI alone, with stabilized inverse probability weighting used to balance baseline patient characteristics. After adjustment, induction chemotherapy only prolonged progression-free survival (PFS) in patients with tumor response to induction, whereas induction chemoimmunotherapy significantly improved PFS relative to both standard treatment and induction chemotherapy, without increasing severe toxic events. No overall survival differences existed across groups. This real-world evidence clarifies the distinct clinical value of different induction strategies and guides personalized treatment selection for this patient population.
Keywords: LA-NSCLC, induction, cCRT, consolidation immunotherapy
Graphical abstract
Highlights
-
•
Induction chemoimmunotherapy extends PFS in unresectable stage III LA-NSCLC after cCRT-I
-
•
Induction chemotherapy shows no overall PFS benefit; responders gain longer PFS
-
•
Induction chemoimmunotherapy improves PFS vs. chemo/no induction with acceptable safety
Health sciences
Introduction
Non-small cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer cases,1 with 30% of NSCLC patients presenting with locally advanced disease (LA-NSCLC) at the time of initial diagnosis; moreover, most of these LA-NSCLC tumors are unresectable.2 Concurrent chemoradiotherapy (cCRT) followed by consolidation durvalumab is the current standard of care (SOC) for NSCLC patients, with the median progression-free survival (mPFS) of 16.8 months and median overall survival (mOS) of 47.5 months, respectively.3,4 The GEMSTONE-301 trial further validated the favorable antitumor activity of consolidation Immune Checkpoint Inhibitors (ICI) following chemoradiotherapy.
It should be noted that according to the PACIFIC trial, the 5-year progression-free survival (PFS) rate was only 33.1%, indicating that nearly 70% of patients with stage III NSCLC experienced uncontrolled disease.4 The survival outcomes of LA-NSCLC remain to be further improved. In addition, 15.4% of patients discontinued treatment due to Treatment-Related Adverse Events (TRAEs). Among patients receiving consolidation durvalumab in this trial, which excluded patients with history of grade 2 or higher pneumonitis before ICI consolidation therapy, the incidences of all-grade and grade 3–4 pneumonitis were 33.9% and 4.2%, respectively. Previous real-world studies indicated that the incidence of grade 3–4 pneumonitis was 14.3%, which limited the administration of consolidation ICI.4 Therefore, it is crucial to explore novel therapeutic strategies to further improve the prognosis and enhance treatment safety in patients with unresectable LA-NSCLC.
The PACIFIC-2 trial attempted to administer concurrent ICIs with cCRT, and the result indicated no efficacy benefit with the increased incidence of Adverse Event (AE).5 Induction therapy can achieve tumor shrinkage; however, whether this strategy can further improve efficacy remains unclear. In the pre-immunotherapy era, the addition of induction chemotherapy had no significant efficacy advantage over cCRT alone. However, certain patient subgroups, such as those with bulky disease or chemotherapy-responsive tumors, may benefit from the addition of induction chemotherapy.6 In the context of consolidation ICI therapy, several retrospective studies on patients with unresectable LA-NSCLC have provided initial insights regarding the efficacy of induction chemoimmunotherapy and delivered promising results.7,8 However, none of these approaches have been utilized to directly investigate the role of induction treatment for the patients receiving cCRT and ICI consolidation therapy. Whether the addition of induction therapy prior to cCRT and immunotherapy consolidation can improve prognosis, and whether induction chemotherapy can achieve non-inferior efficacy compared with chemoimmunotherapy induction while offering superior safety profiles, remain to be further investigated.
Here, we conducted a retrospective study to evaluate the efficacy and safety of induction therapy administered prior to cCRT in patients with unresectable LA-NSCLC.
Results
Results of the modified intent-to-treat population
A total of 202 patients were enrolled and formed the overall modified intent-to-treat (mITT) population. Survival analysis of the mITT cohort showed that the I + C-cCRT-I group had superior PFS relative to the C-cCRT-I (p = 0.023) and non-induction group (p = 0.007). No significant difference of overall survival (OS) was observed between different groups. In line with the design of landmark trials such as PACIFIC, patients who developed progressive disease during induction therapy or did not receive subsequent immune consolidation were excluded from the primary survival analysis. Baseline characteristics and PFS and OS curves of all patients in the mITT population are provided in the supplemental information (Table S1; Figure S1), and the detailed patient screening process is shown in Figure S1.
Baseline characteristics of included patients
A total of 168 eligible patients with LA-NSCLC who had completed the full scheduled treatment course were enrolled. Among these patients, 61 patients (36.3%) received cCRT and subsequent immunotherapy (non-induction group). The other 107 patients (63.7%) received induction therapy, with 36 patients undergoing induction chemoimmunotherapy (I+C-cCRT-I group) and 71 receiving induction chemotherapy (C-cCRT-I group). The baseline characteristics of the patients, including age, sex, Karnofsky Performance Status (KPS), tumor stage, tumor histology type, smoking history, and bulkiness (defined as primary tumor ≥ 5 cm in greatest dimension or metastatic lymph nodes ≥ 2 cm in shortest diameter).6,9,10 PD-L1 expression was evaluated using pretreatment biopsy specimens with the 22C3 or SP263 antibody clones, and the tumor proportion score (TPS) was used for pathological interpretation. Detailed radiotherapy parameters, including V5, V20, V30, and mean lung dose (MLD) of the three groups are shown in Table S2, with no significant differences between groups.
Survival outcomes between induction and non-induction group
Based on the data cutoff date in April 2025, the median follow-up time was 35.2 months (95% confidence interval [CI]: 33.8–38.4). The median PFS1 of all included patients was 29.8 months (95% CI: 26.7–32.9), with the PFS2 of 33.5 months (95% CI: 29.5–37.4).
Baseline characteristics of patients in the induction group and non-induction group are shown in Table 1. The mPFS1 was significantly longer in the induction group than in the non-induction group (32.0 vs. 23.3 months; Hazard Ratio (HR): 0.61, 95% CI: 0.39–0.95, p = 0.029) (Figure 1B), and patients in the induction group had significantly higher 1-, 2-, and 3-year PFS1 rates (88.8%, 78.4%, and 72.6%, respectively) compared with the non-induction group (81.9%, 61.9%, and 47.8%, respectively). Furthermore, PFS2 was also significantly longer in the induction group compared with that in the non-induction group (36.9 vs. 25.1 months, p = 0.011) (Figure 1C).
Table 1.
Baseline demographic and clinical characteristics of the patients
| Variables | Total (n = 168) | Non-induction group (n = 61) | Induction group (n = 107) | p |
|---|---|---|---|---|
| Age (years), n (%) | 0.224 | |||
| <65 | 117 (69.6) | 39 (63.9) | 78 (72.9) | – |
| ≥65 | 51 (30.4) | 22 (36.1) | 29 (27.1) | – |
| Sex, n (%) | 0.985 | |||
| Female | 18 (10.7) | 6 (9.8) | 12 (12.2) | – |
| Male | 150 (89.3) | 55 (90.2) | 95 (88.8) | – |
| KPS, n (%) | 0.566 | |||
| <90 | 64 (38.1) | 21 (34.4) | 43 (40.2) | – |
| ≥90 | 104 (61.9) | 40 (65.6) | 64 (59.8) | – |
| Smoking, n (%) | 0.137 | |||
| Yes | 85 (50.6) | 36 (59.0) | 49 (45.8) | – |
| No | 83 (49.4) | 25 (41.0) | 58 (54.2) | – |
| Histology, n (%) | 0.419 | |||
| Adenocarcinoma | 49 (29.2) | 15 (24.6) | 34 (31.8) | – |
| Squamous | 119 (70.8) | 46 (75.4) | 73 (68.2) | – |
| Stage, n (%) | 0.134 | |||
| IIIA | 77 (45.8) | 24 (39.3) | 53 (49.5) | – |
| IIIB | 66 (39.3) | 30 (49.2) | 36 (33.6) | – |
| IIIC | 25 (14.9) | 7 (11.5) | 18 (16.9) | – |
| Bulky, n (%) | 0.012a | |||
| Yes | 71 (42.3) | 18 (25.4) | 43 (44.3) | – |
| No | 97 (57.7) | 53 (74.6) | 54 (55.7) | – |
| PD-L1 expression, n (%) | 0.862 | |||
| <1% | 28 (16.7) | 11 (18.1) | 17 (15.9) | – |
| 1%–49% | 19 (11.3) | 8 (13.1) | 11 (10.3) | – |
| ≥50% | 20 (11.9) | 6 (9.8) | 14 (13.1) | – |
| Unknown | 101(60.1) | 36 (59.0) | 65 (60.7) | – |
Italicized results denote statistical significance with P < 0.05.
p values were calculated using Wald tests in logistic regression models.
Figure 1.
Study design overview and comparison of PFS and OS between patients with and without induction therapy
(A) The scheme of the study design. The tumor response was assessed pre-chemoradiotherapy (pre-CRT).
(B) PFS1 between the induction and non-induction groups.
(C) PFS2 between the induction and non-induction groups.
(D) OS between the induction and non-induction groups.
Abbreviations: PFS, progression-free survival; OS, overall survival.
In terms of OS, the mOS was not reached (NR) in both cohorts due to the limited follow-up time, and no significant differences were observed between the two groups (HR: 0.83, 95% CI: 0.41–1.69, p = 0.61) (Figure 1D). However, the 1-, 2-, and 3-year OS rates were significantly higher in the induction group (98.1%, 91.5%, and 80.3%, respectively) than in the non-induction group (96.7%, 88.5%, and 78.7%, respectively).
Comparison of survival outcomes between different induction regimens
Survival outcomes before sIPTW
With the median follow-up of 37.5 months, patients in the I +C-cCRT-I group demonstrated a significant improvement in mPFS1 compared with the C-cCRT-I group (NR vs. 29.0 months; HR: 0.39, 95% CI: 0.27–0.87, p = 0.015) (Figure 2A) and non-induction group (NR vs. 23.3 months; HR: 0.39, 95% CI: 0.22–0.70, p = 0.001) (Figure 2D). Similarly, the mPFS2 of patients in I +C-cCRT-I group was also longer than that in the C-cCRT-I group (NR vs. 32.7 months; HR: 0.47, 95% CI: 0.26–0.84, p = 0.01) (Figure 2B) and non-induction group (NR vs.25.2 months; HR: 0.36, 95% CI: 0.21–0.65, p = 0.001) (Figure 2E). Compared with the non-induction group, the C-cCRT-I group demonstrated slightly longer mPFS1 (29.0 vs. 23.3 months; HR: 0.79, 95% CI: 0.51–1.22, p = 0.29) (Figure 2G) and mPFS2 (32.7 vs. 25.2 months; HR: 0.74, 95% CI: 0.47–1.15, p = 0.18) (Figure 2H), though these differences were not statistically significant.
Figure 2.
Before the sIPTW, PFS1, PFS2, and OS between the I+C-cCRT-I and C-cCRT-I groups
I+C-cCRT-I and C-cCRT-I (A‑C), I+C-cCRT-I and non-induction group (D‑F), and C-cCRT-I and non-induction group (G and H).
(A, D, and G) PFS1, (B, E, and H) PFS2, and (C, F, and I) OS. Abbreviations: PFS, progression-free survival; OS, overall survival; I + C, chemoimmunotherapy; C, chemotherapy.
With respect to OS, the 3-year OS rate was 85.0% in the I+C-cCRT-I group compared with 77.9% and 78.7% in the C-cCRT-I and non-induction groups (p = 0.23 and 0.24, respectively) (Figures 2C and 2F); no statistically significant difference was observed in OS between patients in the C-cCRT-I and the non-induction group (NR vs. NR; HR: 0.98, 95% CI: 0.48–2.03, p = 0.96) (Figure 2I).
Survival outcomes after sIPTW
Since younger patients and those with a large tumor burden were more likely to receive induction therapy, we conducted stabilized inverse probability of treatment weighting (sIPTW) . After weighting, the baseline characteristics of the three groups achieved balance (Table 2), and there were 36, 71, and 60.6 patients included in the I+C-cCRT-I, C-cCRT-I, and non-induction groups. The I+C-cCRT-I group exhibited a significant improvement in mPFS1 compared with the C-cCRT-I group (NR vs. 30.8 months; HR: 0.39, 95% CI: 0.167–0.910, p = 0.029) (Figure 3A) and the non-induction group (NR vs. 20.8 months; HR: 0.273, 95% CI:0.13–0.56, p = 0.004) (Figure 3D). The I+C-cCRT-I group also achieved superior PFS2 relative to the C-cCRT-I group (NR vs. 34.6 months; HR: 0.39, 95% CI: 0.17–0.90, p = 0.028) (Figure 3B) and the non-induction group (NR vs. 23.3 months; HR: 0.24, 95% CI: 0.12–0.50, p = 0.002) (Figure 3E). The C-cCRT-I group exhibited longer PFS1 (30.8 vs. 20.8 months; HR: 0.70, 95% CI: 0.45–1.08, p = 0.120) and PFS2 (34.6 vs. 23.3 months; HR: 0.63, 95% CI: 0.41–0.97, p = 0.048) than the non-induction group (Figures 3G and 3H). No statistically significant difference in OS was observed among the three groups (Figures 3G, 3F, and 3I).
Table 2.
Baseline demographic and clinical characteristics of the three groups before and after sIPTW
| Subgroup | Before sIPTW |
After sIPTW |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ICI+chemo induc group (n = 36) | Chemo induc group (n = 71) | Non-indcution group (n = 61) | SMD | p | ICI+chemo induc group (n = 36) | Chemo induc group (n = 71) | Non-indcution group (n = 60.6) | SMD | p | |
| Age, mean ± SD/SE | 58.61 ± 7.38 | 60.28 ± 7.49 | 63.11 ± 6.45 | 0.650 | 0.007a | 60.16 ± 6.65 | 60.83 ± 7.26 | 60.85 ± 7.00 | 0.101 | 0.876 |
| Sex, n (%) | 0.247 | 0.414 | 0.037 | 0.913 | ||||||
| Female | 6 (16.7) | 6 (8.5) | 6 (9.8) | – | – | 4.2 (11.8) | 7.8 (11.1) | 7.4 (12.3) | – | – |
| Male | 30 (88.3) | 65 (91.5) | 55 (90.2) | – | – | 31.8 (88.2) | 63.2 (88.9) | 53.2 (87.7) | – | – |
| ECOG, n (%) | 0.287 | 0.461 | 0.058 | 0.524 | ||||||
| 0 | 20 (55.6) | 48 (67.6) | 40 (65.6) | – | – | 24.6 (68.5) | 45.6 (64.3) | 41.3 (68.2) | – | – |
| 1 | 15 (41.7) | 23 (32.4) | 19 (31.1) | – | – | 10.9 (30.2) | 25.4 (35.7) | 17.2 (28.4) | – | – |
| 2 | 1 (2.8) | 0 (0) | 2 (3.3) | – | – | 0.5 (1.3) | 0.00 (0.0) | 2.1 (3.4%) | – | – |
| Smoking, n (%) | 0.406 | 0.153 | 0.125 | 0.802 | ||||||
| Yes | 14 (38.9) | 35 (49.3) | 36 (59.0) | – | – | 19.9 (55.3) | 34.8 (49.0) | 29.8 (49.1) | – | – |
| No | 22 (61.1) | 36 (50.7) | 25 (41.0) | – | – | 16.1 (44.7) | 36.2 (51.0) | 30.8 (50.9) | – | – |
| Histology, n (%) | 0.172 | 0.603 | 0.070 | 0.995 | ||||||
| Adenocarcinoma | 11 (30.6) | 23 (32.4) | 15 (24.6) | – | – | 10.8 (30.0) | 20.7 (29.1) | 19.2 (31.7) | – | – |
| Squamous | 25 (69.4) | 48 (67.6) | 46 (75.4) | – | – | 25.2 (70.0) | 50.3 (70.9) | 41.4 (68.3) | – | – |
| Stage, n (%) | 0.252 | 0.128 | 0.182 | 0.134 | ||||||
| IIIA | 19 (52.8) | 34 (47.9) | 24 (39.3) | – | – | 17.5 (48.6) | 36.0 (50.7) | 23.1 (38.1) | – | – |
| IIIB | 14 (38.9) | 22 (31.0) | 30 (49.2) | – | – | 15.1 (41.8) | 21.4 (30.1) | 30.9 (50.9) | – | – |
| IIIC | 3 (8.3) | 15 (21.1) | 7 (11.5) | – | – | 3.4 (9.6) | 13.7 (19.2) | 6.7 (11.0) | – | – |
| Bulky, n (%) | 0.481 | 0.037a | 0.105 | 0.865 | ||||||
| Yes | 19 (52.8) | 34 (47.9) | 18 (29.5) | – | – | 13.3 (37.1) | 30.0 (42.2) | 23.8 (39.3) | – | – |
| No | 17 (47.2) | 37 (52.1) | 43 (70.4) | – | – | 22.7 (62.9) | 41.0 (57.8) | 36.8 (60.7) | – | – |
Bold values can be changed to italics to denote statistical significance with P < 0.05.
Data are represented as mean ± SD; ∗∗p values were calculated using Wald tests in logistic regression models.
Figure 3.
After the sIPTW, PFS1, PFS2, and OS between the I+C-cCRT-I and C-cCRT-I groups
I+C-cCRT-I and non-induction group (A‑C), I+C-cCRT-I and C-cCRT-I (D‑F), and C-cCRT-I and non-induction group (G and H).
(A, D, and G) PFS1, (B, E, and H) PFS2, and (C, F, and I) OS. Abbreviations: PFS, progression-free survival; OS, overall survival; I + C, chemoimmunotherapy; C, chemotherapy.
As shown in Tables S3 and S4, univariate and multivariate Cox regression analyses identified I+C-cCRT-I as an independent prognostic factor for PFS1 before sIPTW (HR: 0.21, 95% CI: 0.09–0.51, p < 0.001) and after sIPTW (HR: 0.25, 95% CI: 0.10–0.65, p = 0.004). C-cCRT-I was also an independent prognostic factor for PFS1 before sIPTW (HR: 0.57, 95% CI: 0.36–0.91, p < 0.020) and after sIPTW (HR: 0.59, 95% CI: 0.36–0.97, p = 0.039). Bulky tumor burden remained an independent adverse prognostic factor for PFS1 both before (HR: 1.62, 95% CI: 1.00–2.60, p = 0.048) and after sIPTW (HR: 1.75, 95% CI: 1.04–2.93, p = 0.035).
Subgroup analyses and supplementary analyses for PD-L1 status
Subgroup analyses for PFS1 demonstrated that the I+C-cCRT group showed significantly better survival outcomes than the non-induction group across most subgroups, including patients aged <65 years, males, those with Eastern Cooperative Oncology Group Performance Status (ECOG PS) 0, smokers, non-smokers, those with adenocarcinoma, stage IIIA/IIIB disease, and those with or without bulky lesions. A favorable trend was observed among patients aged ≥65 years, females, those with ECOG PS 1, those with squamous cell carcinoma, and those receiving PD-1 inhibitors, although the differences did not reach statistical significance (Figure S3). Subgroup analyses for PFS2 and OS are presented in Figures S4 and S5.
To address the high proportion of unknown PD-L1 status in the overall cohort, we performed supplementary univariate and multivariate Cox regression analyses after excluding patients with unavailable PD-L1 data. The detailed results are summarized in the supplementary table. Consistent with the primary analysis, the I+C-cCRT-I group remained significantly associated with improved survival compared with the non-induction group (HR: 0.20, 95% CI: 0.05–0.73, p = 0.015). ECOG performance status, clinical stage, and bulky were identified as independent prognostic factors. Notably, no significant correlation was observed between PD-L1 expression level and patient survival outcomes in this subset (Table S5).
Tumor response
We focused on the objective response rate (ORR) after induction therapy; the I+C-cCRT-I group showed a significantly higher induction ORR compared with the C-cCRT-I group (69.4% vs. 42.3%, p = 0.008) (Figure 4A). We also found that compared with non-responders, patients with tumor response had superior PFS1 (NR vs. 26.0 months; HR: 0.34, 95% CI: 0.17–0.67, p = 0.002) and PFS2 (NR vs. 28.0 months; HR: 0.35, 95% CI: 0.18–0.68, p = 0.002) (Figures 4B and 4C). Patients with tumor response showed a numerically higher but statistically non-significant OS compared with non-responders (NR vs. NR months; HR: 0.68, 95% CI 0.26–1.83, p = 0.45) (Figure 4D).
Figure 4.
Tumor responses during induction therapy and their impacts on survival outcomes
(A) Waterfall chart depicting best tumor responses for 36 patients in the I+C-cCRT-I group and 71 patients in the C-cCRT-I group before cCRT.
(B–D) PFS1, PFS2, and OS between patients with response and non-response to induction chemotherapy.
(B) PFS1, (C) PFS2, and (D) OS.
Adverse events
TRAEs occurred in 36 patients (100%), 70 patients (98.6%), and 59 patients (96.7%) in the I+C-cCRT-I, C-cCRT-I, and non-induction groups, respectively (Table S6). The termination due to TRAEs occurred in 24 patients (34.4%) in the C-cCRT-I group, 16 patients (26.2%) in the non-induction group, and 16 patients (44.4%) in the I+C-cCRT-I group. Hematological toxicity was the most common AE among all three groups. Notably, the I+C-cCRT-I group predominantly showed grade 1–2 lymphopenia, whereas the C-cCRT-I and non-induction groups mainly had grade ≥3 lymphopenia.
The incidence of grade ≥2 pneumonitis within or more than 6 months after cCRT was marginally higher in the I+C-cCRT-I group than in the C-cCRT-I and non-induction groups (31.6% vs. 26% vs. 20%, respectively). The long-term pneumonitis incidence showed a comparable trend across the three groups (53% vs. 41% vs. 38%, respectively), with the incidence of grade ≥3 pneumonitis recorded as 17%, 6%, and 8%, respectively. Additionally, there was no recorded incidence of grade 5 TRAEs, except for one case of grade 5 pneumonitis in the non-induction group.
Discussion
Our study confirmed the clinical benefit of the induction therapy in patients with LA-NSCLC undergoing cCRT and consolidation ICI. Further investigation revealed that induction with ICI—rather than chemotherapy—yielded a significant improvement in PFS, with no concomitant increase in AE incidence. To our knowledge, this study is the first retrospective comparative analysis to investigate the clinical benefit of induction chemotherapy with or without ICI for LA-NSCLC patients receiving cCRT and consolidation ICI.
In the pre-immunotherapy era, several studies have evaluated the outcomes of induction chemotherapy prior to cCRT in patients with unresectable NSCLC. A randomized phase 3 trial (CALGB 39801) conducted in 2007 compared survival and toxicity between patients with stage III NSCLC who received cCRT alone and those who received induction chemotherapy followed by cCRT. However, there was no statistically significant difference in survival between the two groups (14 months vs. 12 months, p = 0.30).11 Despite these disappointing findings, it is noteworthy that the HeCOG phase 2 trial performed during the same period demonstrated significantly improved survival among patients who achieved a tumor response after induction chemotherapy followed by cCRT.12 In addition, a retrospective study that included 105 patients with bulky unresectable NSCLC demonstrated the improved PFS for the addition of induction chemotherapy compared with cCRT alone (12.6 vs. 9.0 months, p = 0.02).6 These results indicate that specific patient populations, such as those with large tumor volumes and chemotherapy-responsive disease, may derive clinical benefits from induction chemotherapy. Therefore, this strategy warrants further investigation in the immunotherapy era.
With the emergence and development of immunotherapy, the PACIFIC trial has revolutionized the treatment landscape of unresectable NSCLC. cCRT followed by consolidation ICI has become the SOC for unresectable LA-NSCLC, leading to a significant improvement in prognosis.3 However, previous retrospective studies have demonstrated that only approximately two-thirds of patients receive cCRT in real-world practice.13 High tumor burden and treatment-related adverse events frequently preclude patients from completing cCRT and subsequent consolidation ICI therapy, potentially compromising treatment outcomes. In this context, induction therapy before cCRT has emerged as an optional strategy in clinical practice.
Several clinical trials have provided preliminary evidence investigating this strategy. The phase 2 single-arm APOLO trial employed definitive cCRT plus induction and consolidation atezolizumab, with an mPFS of 20.8 months, a 12-month PFS rate of 68.4%, an 18-month PFS rate of 60.5%, and a 12-month OS rate of 86.8%.14 Another phase 2 single-arm prospective study (AFT-16) investigated induction atezolizumab before cCRT, followed by two cycles of consolidation chemotherapy and consolidation atezolizumab every 21 days for up to 1 year. Among the 62 enrolled patients, outcomes measured from cCRT completion resulted in an mPFS of 30.0 months, a 24-month PFS rate of 66.3%, and a 24-month OS rate of 73.7%.15 Although direct comparisons are not feasible, the PFS values observed in these studies are all longer than the 16.8 months reported in the PACIFIC trial. These findings support the robust antitumor potential of administering induction therapy prior to radiotherapy. However, no clinical trial has yet directly compared the strategy of induction therapy followed by cCRT and consolidation immunotherapy with the current SOC.
In the present retrospective study, we confirmed the clinical benefits associated with the addition of induction therapy prior to cCRT and consolidation ICI for the first time. This favorable outcome may be attributed to the ability of systemic induction therapy to reduce local tumor burden and the risk of micrometastases before radiotherapy, thereby prolonging PFS. However, it is noteworthy that this clinical benefit appears to be derived from combined chemotherapy and ICIs as induction therapy rather than induction chemotherapy alone. Previous studies have demonstrated a strong correlation between large baseline tumor volume prior to radiotherapy and poor survival outcomes following radiotherapy in patients with unresectable NSCLC.16 We found that the ORR after the induction phase in the I+C-cCRT-I group was significantly higher than that in the C-cCRT-I group. Similar conclusions can be drawn in resectable NSCLC. The AEGEAN,17 NADIM II,18 and KEYNOTE-67119 trials have consistently demonstrated that neoadjuvant chemoimmunotherapy, compared with chemotherapy alone, confers favorable outcomes in terms of event-free survival (EFS) and OS. Compared with chemotherapy alone as induction treatment, induction chemoimmunotherapy significantly improved ORR and substantially reduced tumor burden, which may have contributed to the prolonged PFS observed.
Our further analysis revealed that patients who achieved tumor response to induction therapy had significantly better PFS than those without response. This finding supports the notion that an improved ORR may translate into meaningful PFS benefits, and this concept is further validated in advanced metastatic NSCLC. The retrospective SPORE study demonstrated that among patients with metastatic NSCLC receiving first-line chemoimmunotherapy, early tumor response exhibited important prognostic value, as patients with greater tumor shrinkage achieved significantly improved PFS and OS.20 This finding strongly supports the hypothesis proposed in the present study.
From the perspective of its mechanism of action, an increase in the ORR indicates more pronounced tumor regression and a substantial reduction in overall tumor burden. A reduction in tumor burden not only decreases the risk of early distant metastasis but also enhances patient tolerance to subsequent cCRT and consolidation ICI, ensuring the full course of treatment is completed as planned. Ultimately, this may translate a short-term tumor response into a long-term survival benefit. At the same time, induction therapy acts as a natural in vivo drug sensitivity test. For patients with a significant response to induction therapy, continuing the regimen ensures treatment continuity. For those with suboptimal efficacy, chemotherapy can be promptly switched or combined with other therapies to implement a personalized plan, thereby increasing the likelihood of response in subsequent treatments.
In addition, for patients with limited financial resources and poor treatment adherence, administering immunotherapy prior to radiotherapy may further highlight its advantages. It might be a challenge for some patients to complete at least 1 year of consolidation ICI after chemoradiotherapy with adequate quality. A study by Guan et al. indicated that for patients with LA-NSCLC receiving chemoradiotherapy, upfront chemoimmunotherapy may achieve non-inferior outcomes to consolidation immunotherapy with fewer cycles of immunotherapy.8 The efficacy of immunotherapy administered prior to radiotherapy may be attributed to a higher number of tumor antigens in larger, intact tumors, which can effectively activate the host immune system, promote robust T cell infiltration, and enhance antitumor immune responses.21
However, the use of immunotherapy should not be indiscriminate. The PACIFIC-2 trial investigated concurrent immunotherapy with radiotherapy but failed to show improved efficacy.5 Although radiotherapy can induce immunogenic cell death in tumor cells and potentiate immune effects, it may also damage immune cells, particularly lymphocytes, within the radiation field.22 Furthermore, radiotherapy targeting regional lymph nodes impairs the activation of dendritic cells in tumor-draining lymph nodes, thereby reducing immune reactivity. These observations further support that immunotherapy should be administered before, rather than concurrently with, radiotherapy.23
In the present study, although the spectra of TRAEs slightly differed due to distinct treatment modalities, the common TRAEs were relatively similar across these groups. There were 8 patients (8/36, 22.2%) in the I+C-cCRT-I group who permanently discontinued subsequent treatment due to intolerance, with pneumonitis as the primary cause of intolerance in 7 of these patients. In the PACIFIC trial, the incidence of all-grade and grade 3–4 pneumonitis was 33.9% and 4.2%, respectively, in patients receiving durvalumab consolidation therapy.4 In the present study, the incidence of grade ≥3 pneumonitis in the I+C-cCRT-I, C-cCRT-I, and non-induction groups was 17%, 6%, and 8%, respectively, which was higher than that of PACIFIC trial. This may be attributed to the exclusion criteria of the PACIFIC trial, wherein patients with a history of grade 2 or higher pneumonitis before ICI consolidation, who were more susceptible to pulmonary toxicity, were excluded. Additionally, the proportion of patients with grade 3–4 lymphopenia was significantly lower in the I+C-cCRT-I than in the C-cCRT-I and non-induction groups (30.6% vs. 54.9% vs. 54.1%, p = 0.038); moreover, lymphocyte count may also be a critical factor influencing patient outcomes.22
Our findings revealed that for unresectable LA-NSCLC, induction chemotherapy administered prior to cCRT and ICI consolidation yielded PFS benefits solely in patients who achieved a response to induction therapy. However, integrating immunotherapy into the induction regimen significantly prolonged PFS in this patient population, with a well-tolerated safety profile.
Limitations of the study
The present study has some limitations. First, as a single-institution retrospective analysis with a relatively small proportion of female patients, the findings may have limited generalizability. Additionally, PD-L1 expression data were missing in the study population. Although subgroup analyses were performed, the limited sample size may also lead to potential outcome bias. Because of the relatively short follow-up period, the OS outcome remains immature, limiting further exploration of the precise effects of different treatment modalities on prognosis.
Resource availability
Lead contact
Further information and requests for resources should be directed to lead contact, Linlin Wang (llwang@sdfmu.edu.cn).
Materials availability
This study did not use any reagents.
Data and code availability
-
•
De-identified patient standardized data reported in this paper will be shared by the lead contact upon request.
-
•
This paper does not report original code.
-
•
Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.
Acknowledgments
This study was supported by National Key Technologies Research and Development Program (grant no. 2022YFC2404605), National Natural Science Foundation of China (grant no. 82172865), Post-Marketing Clinical Research Special Project on Innovative Drugs (grant no. WKZX2023Cx020012), Natural Science Foundation of Shandong Province (grant nos. ZR2023LZL002, ZR2024MH007, and ZR2021LZL009), Collaborative Academic Innovation Project of Shandong Cancer Hospital (grant no. ZF002), and Noncommunicable Chronic Diseases-National Science and Technology Major Project (grant no. 2024ZD0519904).
Author contributions
S.W. and X.L. designed and supervised the study. Q.Z. and X.Z. were responsible for data acquisition, while Y.M. performed the statistical analysis. The initial manuscript was drafted by Y.M. and subsequently critically revised by J.Z. and L.W. for important intellectual content. All authors reviewed and approved the final manuscript.
Declaration of interests
The authors declare no competing interests.
STAR★Methods
Key resources table
| REAGENT or RESOURCE | SOURCE | IDENTIFIER |
|---|---|---|
| Software and algorithms | ||
| SPSS Statistics | IBM, USA | Version 26.0 |
| R programming language | R Foundation for Statistical Computing | Version 4.2.1 |
| GraphPad Prism | GraphPad Software, USA | Version 10.0 |
Experimental model and study participant details
This retrospective study included patients with histologically confirmed unresectable LA-NSCLC who received induction therapy or not followed by cCRT and ICI consolidation between July 2019 and April 2023 at a hospital. Whether the tumor is resectable or not was confirmed by the multidisciplinary team at our center, including surgeons, oncologists, and radiologists. Eligible patients either received cCRT and consolidation ICI with or without prior induction therapy. The exclusion criteria were as follows: (1) patients harboring epidermal growth factor receptor (EGFR) or anaplastic lymphoma kinase (ALK) mutations; (2) patients who underwent surgical resection or palliative treatment; (3) patients treated with sequential chemoradiotherapy or concurrent ICIs during cCRT.
All included patients received cCRT and consolidation ICI, and were divided into induction group and non-induction group based on whether received induction therapy. Patients in the non-induction group received cCRT and consolidation ICI without induction treatment, and patients in the induction group received either at least 1 cycle of ICI plus chemotherapy or chemotherapy alone as induction therapy, followed by cCRT and consolidation ICI. The induction group was further subdivided into the induction chemoimmunotherapy group (I+C-cCRT-I group) and the induction chemotherapy group (C-cCRT-I group) (Figure 1A ).
All participants enrolled in this retrospective study were East Asian Han Chinese patients, with a mean cohort age of 60.28 ± 7.49 years. The study population consisted of 150 male patients (90.2%) and 18 female patients (9.8%), and no gender-diverse, transgender, or pediatric individuals were included. Biological sex data of all patients were systematically collected and analyzed to explore sex-based differences in survival outcomes and treatment responses. Subgroup analysis demonstrated a trend that male patients tended to achieve more prominent therapeutic benefits from induction therapy. However, several notable limitations exist for this sex-stratified analysis. The cohort displayed severe imbalance in sex distribution, with female patients accounting for merely 9.8% of all enrolled cases. The small sample size of the female subgroup resulted in inadequate statistical power, precluding comprehensive and robust assessment of survival disparities associated with biological sex. Furthermore, only biological sex was recorded in clinical data, while relevant social gender indicators were unavailable, restricting further exploration of potential gender-related influences on treatment efficacy and prognosis.
This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Ethics Committee of the Shandong Cancer Hospital and Institute (Approval number: SDTHEC202509090). As a retrospective study, informed consent was not required for this research.
Method details
Treatment strategy
All included patients received cCRT and consolidation ICI, and were divided into induction group and non-induction group based on whether received induction therapy. Patients in the non-induction group received cCRT and consolidation ICI without induction treatment, and patients in the induction group received either at least 1 cycle of ICI plus chemotherapy or chemotherapy alone as induction therapy, followed by cCRT and consolidation ICI. The induction group was further subdivided into the induction chemoimmunotherapy group (I+C-cCRT-I group) and the induction chemotherapy group (C-cCRT-I group) (Figure 1A ). The ICIs used in induction and consolidation included monoclonal antibodies targeting the programmed cell death-1 (PD-1)/programmed death-ligand 1 (PD-L1) pathway. For patients with squamous cell carcinoma, induction and concurrent chemotherapy included carboplatin or cisplatin combined specifically with paclitaxel, whereas patients with adenocarcinoma were treated with carboplatin or cisplatin plus pemetrexed. cCRT was delivered using intensity-modulated radiation therapy (IMRT). The gross tumor volume of the primary lesion (GTVp) was delineated on post-induction imaging to encompass the residual tumor. The clinical target volume of the primary lesion (CTVp) was generated by applying an 8-mm margin for adenocarcinomas and a 6-mm margin for squamous cell carcinomas. For lymph nodes, the gross tumor volume (GTVn) was defined as the nodal regions initially involved at diagnosis, and the corresponding clinical target volume (CTV) was generated by applying a 5-mm margin. The total radiation dose was 60 Gy ± 10% (54–66 Gy), with at least 1 cycle of chemotherapy administered concurrently.
Quantification and statistical analysis
Clinical outcome measurement and toxicity evaluation
Baseline patient characteristics, clinical manifestations, treatment protocols, and imaging data were collected systematically from the medical records of the patients. All patients underwent routine follow-up. The tumor response was evaluated by chest computed tomography (CT) according to RECIST criteria version 1.1. TRAEs were assessed based on the Common Terminology Criteria for Adverse Events (version 5.0). The PFS1 was defined from the end of cCRT to the date of the first documented event of disease progression or death without progression, which is consistent with the definition of PFS in the PACIFIC trial. PFS2 was defined as the time from the initiation of the first treatment to the date of the first documented event of disease progression or death without progression. OS was defined as the time from initial treatment to death from all causes. The objective response rate (ORR), defined as complete response (CR) plus partial response (PR) by RECIST (version 1.1), was also assessed after induction therapy (induction ORR) (Figure 1A).
Statistical analysis
Baseline characteristics and tumor features were compared between different groups using the chi-square test. Multinomial logistic regression was used to calculate the propensity score for each patient according to the treatment received. All covariates were measured at baseline. Based on the propensity scores, stabilized inverse probability of treatment weighting (sIPTW) was applied to balance covariate distributions across the three groups. Stabilized inverse probability-weighted Cox proportional hazards models were constructed for PFS and OS, and robust standard errors were used to account for the weighting effect. SMD were calculated to evaluate covariate balance before and after weighting. A two-sided p value < 0.05 was considered statistically significant. All statistical analyses and figure plotting were performed using SPSS, R, and GraphPad Prism.
Footnotes
Supplemental information can be found online at https://doi.org/10.1016/j.isci.2026.117034.
Contributor Information
Jingze Zhang, Email: artcher150095@163.com.
Linlin Wang, Email: llwang@sdfmu.edu.cn.
Supplemental information
References
- 1.Bray F., Laversanne M., Sung H., Ferlay J., Siegel R.L., Soerjomataram I., Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 2024;74:229–263. doi: 10.3322/caac.21834. [DOI] [PubMed] [Google Scholar]
- 2.Herbst R.S., Morgensztern D., Boshoff C. The biology and management of non-small cell lung cancer. Nature. 2018;553:446–454. doi: 10.1038/nature25183. [DOI] [PubMed] [Google Scholar]
- 3.Antonia S.J., Villegas A., Daniel D., Vicente D., Murakami S., Hui R., Kurata T., Chiappori A., Lee K.H., de Wit M., et al. Overall Survival with Durvalumab after Chemoradiotherapy in Stage III NSCLC. N. Engl. J. Med. 2018;379:2342–2350. doi: 10.1056/NEJMoa1809697. [DOI] [PubMed] [Google Scholar]
- 4.Spigel D.R., Faivre-Finn C., Gray J.E., Vicente D., Planchard D., Paz-Ares L., Vansteenkiste J.F., Garassino M.C., Hui R., Quantin X., et al. Five-Year Survival Outcomes From the PACIFIC Trial: Durvalumab After Chemoradiotherapy in Stage III Non-Small-Cell Lung Cancer. J. Clin. Oncol. 2022;40:1301–1311. doi: 10.1200/jco.21.01308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bradley J.D., Sugawara S., Lee K.H.H., Ostoros G., Demirkazik A., Zemanova M., Sriuranpong V., Gelatti A., Menezes J., Zurawski B., et al. LBA1 Durvalumab in combination with chemoradiotherapy for patients with unresectable stage III NSCLC: Final results from PACIFIC-2. ESMO Open. 2024;9 doi: 10.1016/j.esmoop.2024.102986. [DOI] [Google Scholar]
- 6.Ahmed I., Ferro A., Baby R., Malhotra J., Cohler A., Langenfeld J., Aisner J., Zou W., Jabbour S.K. Modern induction chemotherapy before chemoradiation for bulky locally-advanced nonsmall cell lung cancer improves survival. J. Cancer Res. Ther. 2016;12:952–958. doi: 10.4103/0973-1482.177214. [DOI] [PubMed] [Google Scholar]
- 7.Fu Z., Dong X., Sun J., Deng W., Zhao Y., Yang D., Jiang L., Chang X., Yu R., Shi A., et al. Induction immunochemotherapy followed by definitive chemoradiotherapy and consolidation immunotherapy for unresectable stage III non-small cell lung cancer: a multi-institutional retrospective cohort study. Front. Immunol. 2025;16 doi: 10.3389/fimmu.2025.1602082. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Guan S., Wang Y., Liu Q., Yan M., Ren K., Wang J., Bi N., Zhao L. Induction chemoimmunotherapy may achieve non-inferior outcomes to consolidation immunotherapy in patients with unresectable stage III NSCLC: a real-world multicenter retrospective study. Front. Immunol. 2025;16 doi: 10.3389/fimmu.2025.1591134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Robinson L.A., Ruckdeschel J.C., Wagner H., Jr., Stevens C.W., American College of Chest Physicians Treatment of non-small cell lung cancer-stage IIIA: ACCP evidence-based clinical practice guidelines (2nd edition) Chest. 2007;132:243s–265s. doi: 10.1378/chest.07-1379. [DOI] [PubMed] [Google Scholar]
- 10.Wang Y., Zhang T., Wang J., Zhou Z., Liu W., Xiao Z., Deng L., Feng Q., Wang X., Lv J., et al. Induction Immune Checkpoint Inhibitors and Chemotherapy Before Definitive Chemoradiation Therapy for Patients With Bulky Unresectable Stage III Non-Small Cell Lung Cancer. Int. J. Radiat. Oncol. Biol. Phys. 2023;116:590–600. doi: 10.1016/j.ijrobp.2022.12.042. [DOI] [PubMed] [Google Scholar]
- 11.Vokes E.E., Herndon J.E., 2nd, Kelley M.J., Cicchetti M.G., Ramnath N., Neill H., Atkins J.N., Watson D.M., Akerley W., Green M.R., Cancer and Leukemia Group B Induction chemotherapy followed by chemoradiotherapy compared with chemoradiotherapy alone for regionally advanced unresectable stage III Non-small-cell lung cancer: Cancer and Leukemia Group B. J. Clin. Oncol. 2007;25:1698–1704. doi: 10.1200/jco.2006.07.3569. [DOI] [PubMed] [Google Scholar]
- 12.Kosmidis P., Fountzilas G., Baka S., Samantas E., Dimopoulos A.M., Gogas H., Skarlos D., Papacostas P., Boukovinas J., Bakogiannis C., et al. Combination chemotherapy with paclitaxel and gemcitabine followed by concurrent chemoradiotherapy in non-operable localized non-small cell lung cancer. A hellenic cooperative oncology group (HeCOG) phase II study. Anticancer Res. 2007;27:4391–4395. [PubMed] [Google Scholar]
- 13.Filippi A.R., Bar J., Chouaid C., Christoph D.C., Field J.K., Fietkau R., Garassino M.C., Garrido P., Haakensen V.D., Kao S., et al. Real-world outcomes with durvalumab after chemoradiotherapy in patients with unresectable stage III NSCLC: interim analysis of overall survival from PACIFIC-R. ESMO Open. 2024;9 doi: 10.1016/j.esmoop.2024.103464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Provencio M., Campos B., Guirado M., Vila L., García Campelo R., Dorta M., Vázquez Estévez S., Ferrández A., Sala M.Á., Ortega A.L., et al. Induction chemo-immunotherapy followed by chemo-radiotherapy and immunotherapy maintenance in stage III NSCLC (APOLO): a phase 2 trial. Nat. Commun. 2025;16 doi: 10.1038/s41467-025-66097-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ross H.J., Kozono D., Wang X.F., Urbanic J.J., Williams T.M., Nelson G.D., Carbone D.P., Chung D., Robb R., Byun W.Y., et al. Atezolizumab Before and After Chemoradiation for Unresectable Stage III Non-Small Cell Lung Cancer: A Phase II Nonrandomized Controlled Trial. JAMA Oncol. 2024;10:1212–1219. doi: 10.1001/jamaoncol.2024.1897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Werner-Wasik M., Swann R.S., Bradley J., Graham M., Emami B., Purdy J., Sause W. Increasing tumor volume is predictive of poor overall and progression-free survival: secondary analysis of the Radiation Therapy Oncology Group 93-11 phase I-II radiation dose-escalation study in patients with inoperable non-small-cell lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 2008;70:385–390. doi: 10.1016/j.ijrobp.2007.06.034. [DOI] [PubMed] [Google Scholar]
- 17.Heymach J.V., Harpole D., Mitsudomi T., Taube J.M., Galffy G., Hochmair M., Winder T., Zukov R., Garbaos G., Gao S., et al. Perioperative Durvalumab for Resectable Non-Small-Cell Lung Cancer. N. Engl. J. Med. 2023;389:1672–1684. doi: 10.1056/NEJMoa2304875. [DOI] [PubMed] [Google Scholar]
- 18.Provencio M., Nadal E., González-Larriba J.L., Martínez-Martí A., Bernabé R., Bosch-Barrera J., Casal-Rubio J., Calvo V., Insa A., Ponce S., et al. Perioperative Nivolumab and Chemotherapy in Stage III Non-Small-Cell Lung Cancer. N. Engl. J. Med. 2023;389:504–513. doi: 10.1056/NEJMoa2215530. [DOI] [PubMed] [Google Scholar]
- 19.Wakelee H., Liberman M., Kato T., Tsuboi M., Lee S.H., Gao S., Chen K.N., Dooms C., Majem M., Eigendorff E., et al. Perioperative Pembrolizumab for Early-Stage Non-Small-Cell Lung Cancer. N. Engl. J. Med. 2023;389:491–503. doi: 10.1056/NEJMoa2302983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Moinard-Butot F., Barbe-Richaud J.B., Lasvergnas J., Ancel J., Justeau G., Le Floch H., Somme L., Chouaid C., Bylicki O., Schott R. Prediction of survival and outcomes with depth of response at 6 months in metastatic non-small-cell lung cancer patients treated with chemotherapy and immunotherapy combination: SPORE trial. ESMO Open. 2026;11 doi: 10.1016/j.esmoop.2025.106042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Forde P.M., Chaft J.E., Smith K.N., Anagnostou V., Cottrell T.R., Hellmann M.D., Zahurak M., Yang S.C., Jones D.R., Broderick S., et al. Neoadjuvant PD-1 Blockade in Resectable Lung Cancer. N. Engl. J. Med. 2018;378:1976–1986. doi: 10.1056/NEJMoa1716078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pike L.R.G., Bang A., Mahal B.A., Taylor A., Krishnan M., Spektor A., Cagney D.N., Aizer A.A., Alexander B.M., Rahma O., et al. The Impact of Radiation Therapy on Lymphocyte Count and Survival in Metastatic Cancer Patients Receiving PD-1 Immune Checkpoint Inhibitors. Int. J. Radiat. Oncol. Biol. Phys. 2019;103:142–151. doi: 10.1016/j.ijrobp.2018.09.010. [DOI] [PubMed] [Google Scholar]
- 23.Telarovic I., Yong C.S.M., Kurz L., Vetrugno I., Reichl S., Fernandez A.S., Cheng H.W., Winkler R., Guckenberger M., Kipar A., et al. Delayed tumor-draining lymph node irradiation preserves the efficacy of combined radiotherapy and immune checkpoint blockade in models of metastatic disease. Nat. Commun. 2024;15:5500. doi: 10.1038/s41467-024-49873-y. [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
-
•
De-identified patient standardized data reported in this paper will be shared by the lead contact upon request.
-
•
This paper does not report original code.
-
•
Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.