Thoracic Radiation on Survival in Patients with Advanced Small Cell Lung Cancer Treated with Chemoimmunotherapy

Introduction

According to the latest data released by CA: A Cancer Journal for Clinicians (2022) global cancer statistics, lung cancer remains the malignancy with both the highest incidence rate and mortality rate worldwide [1]. Small cell lung cancer (SCLC), representing about 15% of lung tumors, is highly aggressive, with unfavorable clinical outcomes, with approximately 70% of SCLC patients having advanced-stage disease [2,3]. Over the previous 20 years, the median survival duration for patients with advanced-stage (AS)-SCLC was 9 to 11 months. Based on the CASPIAN and IMpower133 trials, combined immunotherapy and chemotherapy are authorized by National Comprehensive Cancer Network guidelines as the primary treatments for AS-SCLC. However, compared to just chemotherapy, a 2-month increase in overall survival (OS) was observed [4–6]. To increase the survival of patients with AS-SCLC, additional potent therapies are urgently required. Recent advances in immunotherapy, particularly immune checkpoint inhibitors (ICIs), have transformed the treatment landscape for AS-SCLC [7]. However, the clinical benefits of ICIs remain limited in many patients, highlighting the need for strategies to enhance their efficacy. Thoracic radiation therapy (TRT) has been considered as a promising approach to potentiate the effects of immunotherapy. These changes promote dendritic cell activation and tumor-specific T-cell priming, thereby enhancing the anti-tumor immune response. This synergistic interaction between TRT and immunotherapy provides a strong biological rationale for combining these modalities to improve outcomes in patients with AS-SCLC [8].

TRT is crucial in the treatment of SCLC. Combined chemoradiotherapy has become the preferred method for treating early-stage SCLC. After first-line treatment, intrathoracic progression is still present in at least 75% of patients with AS-SCLC [9]. According to the CREST trial, Radiation Therapy Oncology Group protocol 0937, and a meta-analysis, thoracic radiotherapy enhances local control and survival of AS-SCLC [10–12]. Through T-cell-dependent cytotoxic processes, programmed death-ligand 1 (PD-L1) inhibitors can improve the effectiveness of radiotherapy by decreasing PD-L1 levels on tumor cells [13]. Recent advances in cancer biology have elucidated the potential synergistic mechanisms between TRT and immunotherapy. TRT can induce immunogenic cell death, releasing tumor-associated antigens that prime anti-tumor T-cell responses. This radiation-induced immunogenic effect is particularly relevant in the context of ICIs, as it can help overcome tumor immune evasion mechanisms. Additionally, TRT has been shown to upregulate PD-L1 expression on tumor cells, potentially enhancing their susceptibility to programmed cell death-1 (4)/PD-L1 inhibitors [14]. Radiation also modulates the tumor microenvironment by promoting T-cell infiltration and enhancing antigen presentation. These mechanisms provide a strong biological rationale for combining TRT with immunotherapy in AS-SCLC, potentially leading to improved systemic tumor control and survival outcomes.

While previous studies have explored the role of TRT in AS-SCLC, most have focused on its use in the context of chemotherapy alone [10–12]. The integration of immunotherapy, particularly ICIs, into the first-line treatment of AS-SCLC has fundamentally changed the therapeutic landscape, yet the potential synergistic effects of TRT and chemoimmunotherapy remain poorly understood [4,5]. The present study uniquely contributes to the field by specifically evaluating the effect of TRT in patients with AS-SCLC receiving modern chemoimmunotherapy regimens. By analyzing real-world data, we provide novel insights into whether TRT can further enhance the survival benefits of chemoimmunotherapy, a question that has not been systematically addressed in prior research. Through our findings, we aim to inform clinical practice and guide future randomized controlled trials designed to optimize the integration of TRT in the chemoimmunotherapy era. Therefore, in this retrospective study from a single center, we aimed to compare outcomes in 172 patients with AS-SCLC treated with chemotherapy and immunotherapy, with and without TRT. We hypothesized that consolidative TRT would improve progression-free survival (PFS) and OS in patients with AS-SCLC receiving first-line chemoimmunotherapy.

Material and Methods

PATIENT SELECTION:

This investigation was conducted with approval by the Ethics Committee of Tianjin Medical University Cancer Institute and Hospital (approval number: bc2022244) and in accordance with the ethical standards of the Declaration of Helsinki. All patients gave consent through an informed consent process that was reviewed by the ethics committee. Altogether, 172 patients with SCLC participated in the present investigation, from October 2018 to March 2022. The inclusion criteria were as follows: (1) patients with pathologically confirmed SCLC aged 18 years or older; (2) patients with diagnosis of advanced stage per the American Joint Committee on Cancer (AJCC) Tumor, Node, Metastasis (TNM) staging system, 8th edition, and the Veterans Administration Lung Group (VALG) staging system [14,15] (ie, stage IV disease or cases in which the tumor or nodal burden was too extensive to be encompassed within a safe radiation plan, including T3–T4 tumors characterized by multiple pulmonary nodules); (3) patients treated with or without consolidated chest radiation therapy after partial response or stable disease following first-line chemotherapy combined with immunotherapy; and (4) an Eastern Cooperative Oncology Group (ECOG) performance status ≤2 (ambulatory and capable of all self-care; and up and about more than 50% of waking hours) [16]. The following exclusion criteria were applied: (1) malignancy in other sites (before or concurrently), (2) no single histological type, and (3) autoimmune disorder.

RADIOTHERAPY:

After receiving chemotherapy plus immunotherapy as the primary treatment, 60 patients underwent TRT. Radiation therapy for most patients was intensity modulated radiotherapy or volumetric modulated arc therapy, while 1 patient was treated with CyberKnife. The gross tumor volume was the bulk of tumor that could be seen on the planning computed tomography scan, including the primary lung lesions and involved lymph nodes. For those patients who had received chemoimmunotherapy before radiotherapy, the primary lesion was outlined based on the post-chemotherapy imaging, while the involved lymph nodes were defined in accordance with the pre-treatment imaging. The clinical target volume was created by expanding the gross tumor volume with a 5-mm 3-dimensional margin, while the planning risk volume was created by a 5-mm expansion of the clinical target volume. Of the 60 patients who underwent TRT, 36 patients received conventional radiotherapy (prescribed dose: 50–63 Gy in 25–30 fractions), 16 received palliative radiotherapy (prescribed dose: 30–45 Gy in 10–15 fractions), 5 received hyperfractionated radiotherapy (45 Gy in 3 weeks in twice-daily fractionation), and 3 received low-dose radiotherapy (prescribed dose: 15–21 Gy in 3–5 fractions). We calculated the equivalent dose in 2 Gy fractions (EQD2), and the median EQD2 dose was 51.53 (16.25–63.53) Gy. The dose constraints for organs-at-risk were as follows: total lung V20 <28%, V30 <20%, V5 <60%, mean lung dose <16 Gy; esophagus Dmax <66 Gy, mean dose <26 Gy; heart V30 <45%, mean dose <20 Gy; and spinal cord Dmax <45 Gy.

CHEMOTHERAPY AND IMMUNOTHERAPY:

The chemotherapy regimens applied were etoposide combined with cisplatin or carboplatin, the dosages and administration of which were strictly in accordance with National Comprehensive Cancer Network guidelines. The drugs were given every 3 weeks for 4 to 6 cycles, depending on the tolerance of the patients. The immunotherapy agents included antibodies against PD-1 and PD-L1, such as nivolumab, pembrolizumab, sintilimab, camrelizumab, serplulimab, tislelizumab, atezolizumab, and durvalumab. Total immunotherapy cycles were 6.5 (range, 1–54) in the whole cohort, 7 (range, 2–33) in the TRT group, and 6 (range, 1–54) in the group that did not receive TRT (non-TRT). All patients received combined chemoimmunotherapy for 4 to 6 cycles initially, and maintenance immunotherapy until disease progression or intolerance. In the TRT group, 58 patients received 4 to 6 cycles of immunotherapy followed by TRT, and 2 patients received concurrent immunotherapy with radiotherapy from the beginning. Stable disease before radiotherapy or partial response was observed in 58 patients with sequential radiotherapy, while 2 patients had disease progression. Salvage therapy, such as chemotherapy, immunotherapy, or anlotinib, was administered to patients with progression after first-line treatment. In addition, palliative radiotherapy at the site of the metastases was given to patients with bone, brain, or adrenal metastases.

STATISTICAL ANALYSIS:

The clinical and demographic features of patients were collected, such as age, sex, ECOG scores, total stage, tumor, node, oligometastases, brain metastases, bone metastases, and liver metastases. The primary and secondary end points of the study were OS and PFS, respectively. OS was described as the time passing between the start of therapy and death or having their last examination. PFS was defined as the time between the start of treatment and the onset of the disease, or death. Mann-Whitney U tests were used for continuous variables, while the Fisher exact or chi-square tests were applied to categorical variables. Furthermore, with the use of univariate and multivariate Cox regression analyses, the specified variables’ effects on OS and PFS were assessed. OS and PFS were evaluated by the Kaplan-Meier survival analysis. Log-rank tests were performed to assess survival differences between the TRT and non-TRT groups.

To reduce potential selection bias between the TRT and non-TRT groups, propensity score matching (PSM) was conducted, with TRT dose as a covariate. The nearest-neighbor approach was used, with a caliper of 0.02 to construct a matched control sample for the PSM study. OS and PFS for the 2 groups were compared after PSM. While PSM helps balance measured confounders between groups, it cannot account for unmeasured confounders, such as differences in performance status, comorbidities, or access to care, which can influence outcomes. Subgroup analysis was performed to identify the beneficial groups of TRT, and the forest plot was used to show the hazard ratio (HR) and 95% confidence interval (CI) for each subgroup. We analyzed the data with SPSS (version 27; IBM Corp, Armonk, NY, USA) and R (version 4.1.3). The main R packages used were survival, survminer, tableone, and forestplot. Statistical significance was attributed to P value less than 0.05.

Results

CLINICAL FEATURES:

The analysis included 172 patients in total who fulfilled the inclusion criterion. The average follow-up was 20.1 months (95% CI: 17.06–23.14). The patients’ baseline features are summarized in Table 1. The use of PSM allowed us to balance baseline characteristics between the TRT and non-TRT groups, reducing the potential for confounding by measured variables. After PSM, the cohorts were well-balanced in terms of all included covariates. This balancing ensured that the observed survival benefits associated with TRT were less likely to be confounded by baseline differences. About 38.37% of the total patient population was above 65 years of age, while the average patient age was 62 years (range 29–79 years). In the present study, 131 patients (76.2%) were men, and 160 (93.0%) had an ECOG performance status of less than 1. Eligible study participants were categorized as those who received TRT (n=60) and those who did not (non-TRT, n=112). According to the AJCC/TNM 8th edition and VALG staging system, there were 8 patients (4.7%) with stage IIIB/IIIC, 61 (35.5%) with stage IVA, and 103 (59.9%) with stage IVB.

SURVIVAL AND PROGNOSIS ANALYSIS:

We selected covariates for the multivariable Cox regression model through a combination of univariate screening (P<0.05) and clinical judgment. This approach ensured the inclusion of statistically significant variables and key clinical predictors. The covariates incorporated into the final model were age (≥65 years), oligometastasis, bone metastasis, liver metastasis, and brain metastasis. Brain metastasis was included based on clinical rationale, as it can substantially influence survival outcomes in patients undergoing radiotherapy, even though the statistical significance in univariate analysis was borderline (P=0.186 for OS; P=0.639 for PFS). Other covariates (age ≥65 years, oligometastasis, bone metastasis, and liver metastasis) were selected based on statistical significance in univariate analysis (P<0.05).

The overall median OS was 19.7 months (95% CI: 16.14–23.32), while PFS was 8.4 months (95% CI: 7.42–9.44). The median OS for the TRT group was 24.2 months (95% CI: 18.33–30.07), and that of the non-TRT group was 15.9 months (95% CI: 10.23–21.57, P=0.006), while the median PFS values were 11.3 months (95% CI: 10.07–12.53) and 6.6 months (95% CI: 5.63–7.57, P<0.001), respectively (Figure 1). On univariate analysis, 4 factors were found to be associated with OS, including TRT, age ≥65 years, bone metastasis, and liver metastasis (Table 2). The factors of TRT, age ≥65 years, bone metastasis, oligometastasis (defined as 1 to 5 metastatic lesions), and liver metastasis were associated with PFS. The outcomes of the multivariate analysis indicated that liver metastasis and TRT were independent predictors for OS, whereas age ≥65 years and TRT were independent predictors for PFS (Table 3).

The 1: 1 PSM, which was used to adjust bone metastasis, yielded 60 patients in each group. After matching by propensity score, none of the relevant variables had a substantial difference between the 2 cohorts. In addition, compared with the non-RT group, the TRT group still had significantly prolonged OS and PFS after PSM (Figure 2). Median OS was 24.2 months (95% CI: 18.33–30.07) in the TRT group and 18.0 months (95% CI: 12.54–23.46, P=0.006) in the non-TRT group. Median PFS was 11.3 months (95% CI: 10.07–12.53) in the TRT group and 6.7 months (95% CI: 4.72–8.74, P<0.001) in the non-TRT group.

SUBGROUP ANALYSES:

We performed subgroup analyses to identify potential beneficial factors for TRT. In addition to the characteristics described in Table 3, we added the variable of response to first-line immunotherapy. Subgroup analysis showed that patients in all subgroups had OS benefit after TRT. According to subgroup analyses of OS, patients without bone metastasis (HR: 0.55, P=0.042) and without brain metastasis (HR: 0.55, P=0.027) appeared to have longer OS (Figure 3). In the subgroup analysis of PFS, most patients who underwent TRT achieved longer PFS, with patients without brain and bone metastases appearing to benefit more (HR: 0.49, P<0.001 and HR: 0.48, P<0.001, respectively; Figure 4).

Discussion

This retrospective study showed the effectiveness of TRT in addition to first-line PD-1/PD-L1-antibody-based chemoimmunotherapy in patients with AS-SCLC. The results indicated improved survival with TRT following immunotherapy. In our study, the non-TRT group had median PFS and OS of 6.6 and 15.9 months, respectively, consistent with earlier findings from the IMPOWER133 trial (12.3 and 5.2 months), CASPIAN trial (12.9 and 5.1 months), and ASTRUM-005 trial (15.9 and 5.7 months) [18,19]. This implies that our results are generalizable and relevant to individuals with AS-SCLC. There have been several exploratory investigations into the safety and efficacy of combining TRT with chemoimmunotherapy in patients with AS-SCLC. For instance, Welsh et al examined the effects of co-treatment using pembrolizumab and thoracic radiation, finding a median follow-up, PFS, and OS of 7.3, 6.1, and 8.4 months, respectively [20]. Welsh et al performed chemotherapy alone before TRT, whereas we conducted chemotherapy in combination with immunotherapy. Xie et al observed that, among patients with AS-SCLC undergoing first-line chemoimmunotherapy following consolidative TRT, the median values for PFS and OS were 8 and 22.7 months, respectively [21]. The study conducted by Xie et al involved a cohort of 45 patients who received radiation therapy, a sample size smaller than that of our own study, which included 60 patients. During the 2023 ASCO Annual Meeting, it was reported that the median PFS for patients receiving SHR-1316 together with chemotherapy followed by TRT as first-line treatment was 7 months (CI: 4.3–9.7) [22]. It is worth noting that although the study was forward-looking, it was conducted as a single-arm study. Our study showed a longer survival than did the above-mentioned study. We think the possible reasons for the survival advantage are as follows. First, our patients were administered higher doses of TRT, with a median EQD2 dose of 51.53 Gy. Second, the patients in our study who had disease progression following first-line treatment received adequate salvage therapy, which may have improved their prognosis. The vast majority of patients in our cohort received sequential TRT and immunotherapy, which may limit the generalizability of our findings to patients receiving concurrent therapy.

Emerging evidence suggests that TRT can enhance immunotherapy efficacy through multiple mechanisms. First, TRT upregulates PD-L1 expression on tumor cells, potentially increasing their susceptibility to PD-1/PD-L1 inhibitors. Second, radiation promotes dendritic cell maturation and antigen presentation, enhancing T-cell activation. Furthermore, immunotherapy can enhance the effects of TRT through several mechanisms. ICIs can reverse radiation-induced T-cell exhaustion, prolonging anti-tumor immune responses [23]. Additionally, by blocking PD-1/PD-L1 interactions, immunotherapy can enhance the abscopal effect of TRT, potentially improving control of non-irradiated metastases [24].

Although PSM effectively balanced observed covariates between treatment groups, several important limitations of this study must be acknowledged. First, as a retrospective analysis, we could not account for unmeasured confounders that can influence treatment selection and outcomes, such as detailed tumor biology and comprehensive comorbidity profiles. Second, PSM relies on the accuracy and completeness of recorded variables, and residual confounding can persist despite our rigorous matching approach. Third, the generalizability of our findings can be limited by the exclusion of unmatched patients during the PSM process. Furthermore, the retrospective nature of this study introduced several inherent limitations. Treatment selection was not randomized, and despite our rigorous statistical adjustments, we cannot exclude the possibility of selection bias. Additionally, the lack of standardized protocols for some aspects of patient management may have introduced variability in outcomes. These limitations underscore the need for prospective validation of our findings in randomized controlled trials.

Advanced age emerged as a significant predictor of poorer outcomes, likely reflecting multiple factors, including decreased treatment tolerance, higher comorbidity burden, and potential age-related differences in tumor biology. The protective effect of TRT aligns with its dual role in local disease control and potential immune modulation. Beyond direct tumor cell killing, TRT may enhance systemic immune responses through different mechanisms, including increased tumor antigen release and improved T-cell infiltration, which is particularly relevant in the context of immunotherapy. This can reflect both the biological aggressiveness of liver metastasis and its potential to create an immunosuppressive microenvironment, possibly limiting the efficacy of systemic therapies. These predictive factors have important clinical implications. They can help identify patients who could derive particular benefit from TRT and those who can require alternative or intensified treatment strategies.

The subgroup analysis showed that all patients had a trend of survival benefit after TRT, and the patients with an absence of brain metastasis or bone metastasis had superior OS and PFS. Slotman et al conducted a reanalysis of the CREST study and discovered that, in patients with bone metastasis, those who received TRT had better PFS than those who did not (HR 1.60, P=0.02), but there were no significant differences in OS [25]. This contrasts with our results. A possible explanation is the immunotherapy context of our study, which requires further investigation. The observed survival benefit of TRT in patients without brain or bone metastases can be explained by several biological mechanisms. First, the absence of these metastases typically indicates lower overall tumor burden, which has been associated with enhanced systemic immune responses. Second, brain and bone metastases are known to create immunosuppressive microenvironments that could potentially limit the abscopal effect of TRT. Third, the blood-brain barrier and unique bone marrow niche may restrict immune cell trafficking and function, potentially reducing the synergy between TRT and immunotherapy in these metastatic sites. These findings highlight the need for further research into the interaction between metastatic patterns and TRT efficacy. Future studies should incorporate comprehensive immune profiling, to better understand how different metastatic sites influence systemic immune responses. Additionally, the development of strategies to overcome the immunosuppressive microenvironment of brain and bone metastases could potentially extend the benefits of TRT to a broader patient population. While our study demonstrated overall benefit of TRT, the identification of non-responders and poor-prognosis subgroups has important clinical implications. For patients with high tumor burden or liver metastases, alternative approaches, such as dose-escalated radiotherapy or combination with novel radiosensitizers, can be warranted. Additionally, the association between baseline inflammatory markers and poor outcomes suggests the potential utility of incorporating these biomarkers into treatment decision-making.

This study had several limitations. First, the study was a retrospective analysis conducted at one institution, and thus affected by potential selection bias. Second, the limited sample size and diversity of treatment could have impacted the outcomes. While our primary analysis showed statistically significant results, we acknowledge the potential limitations of our modeling approach. The possibility of residual confounding remains, particularly from unmeasured variables, such as molecular biomarkers or detailed treatment toxicity profiles. These limitations highlight the need for prospective validation of our findings. Furthermore, although our study provides valuable insights into the combination of TRT and immunotherapy, it is important to note that most patients received sequential therapy. This limitation precludes definitive conclusions about the optimal timing of TRT relative to immunotherapy administration. Preclinical studies suggest that concurrent administration can enhance the abscopal effect through radiation-induced immunogenic cell death coinciding with immune checkpoint inhibition. However, the potential for increased toxicity with concurrent therapy requires careful consideration. While our subgroup analyses suggest consistent TRT benefits across response categories, residual confounding due to unmeasured selection factors, such as patient fitness and clinician bias, cannot be excluded. Prospective trials with protocol-defined TRT criteria are needed to validate these findings. While our results suggest potential benefits of TRT in patients receiving chemoimmunotherapy, they cannot be generalized to concurrent administration strategies. Future studies should specifically investigate the timing of TRT relative to immunotherapy, including direct comparisons of concurrent versus sequential approaches. The retrospective design and the potential for unmeasured confounders, despite the use of PSM and multivariate adjustments, may have interfered with the results. Randomized controlled trials are required to validate the findings and to define the optimal integration of TRT in the chemoimmunotherapy era.

Conclusions

In conclusion, our study results suggested that combining TRT with first-line chemoimmunotherapy might improve survival outcomes in patients with AS-SCLC. However, these findings should be interpreted with caution, due to the retrospective design and inherent limitations, including potential selection bias, single-institution nature of the data, and lack of randomization. Therefore, our findings call for large prospective, randomized controlled trials to further elucidate the role of TRT in patients with AS-SCLC receiving chemoimmunotherapy.