Abstract
Background/Aim: Many patients with advanced lung cancer develop brain metastasis (BM); however, few reports confirming the efficacy of immune checkpoint inhibitors (ICIs) plus chemotherapy in non-small cell lung cancer (NSCLC) patients with symptomatic BM have been published. Therefore, we retrospectively evaluated the effects of chemoimmunotherapy in NSCLC patients who did or did not receive prior brain radiotherapy. Patients and Methods: A total of 103 patients with advanced NSCLC who received ICIs plus chemotherapy at our hospital from January 2019 to July 2021 were retrospectively enrolled. Results: Patients with BM tended to have shorter progression-free survival (PFS) and overall survival (OS) compared with patients without BM. The maximum size of BM and the proportion of patients with symptomatic BM were greater among patients who received brain radiotherapy before chemoimmunotherapy. However, patients who received prior brain radiotherapy had better PFS and OS compared with patients who did not receive prior brain radiotherapy. Conclusion: Patients who received prior brain radiotherapy experienced a superior therapeutic benefit of ICIs plus chemotherapy, including those with larger and more symptomatic BM.
Approximately one-third of patients with advanced non-small cell lung cancer (NSCLC) develop brain metastasis (BM) (1). With a poor prognosis, BM is one of the leading causes of cancer-related death. Currently, therapies such as surgery, radiotherapy, chemotherapy, targeted therapies, and immunotherapy are performed for BM. Among these, the efficacy of immune checkpoint inhibitors (ICIs) has been documented by several research groups (2).
Previous studies of programmed cell death-1 (PD-1)/programmed cell death-ligand-1 (PD-L1) inhibitors in lung cancer have demonstrated that PD-L1 expression is a predictive biomarker for ICI efficacy (3, 4). Although heterogeneity in PD-L1 expression is sometimes observed between primary NSCLC tumors and BM in some patients, it is concordant in most patients (5). In fact, Goldberg et al. reported that 29.7% of patients with NSCLC BM with a PD-L1 tumor proportion score (TPS) ≥1% achieved a BM response after receiving ICIs, while none of the patients with a PD-L1 TPS <1% experienced a BM response (6), suggesting that PD-L1 expression in primary tumor cells is associated with BM response to ICIs.
Currently, PD-1/PD-L1 inhibitors combined with chemotherapy are used to treat patients with NSCLC (7-9). In a pooled analysis of the KEYNOTE-189, −021, and −407 studies, pembrolizumab plus chemotherapy prolonged progression-free survival (PFS) and overall survival (OS) compared to standard chemotherapy in patients with or without BM across all PD-L1 expression subgroups (10). However, a small proportion of patients received brain radiotherapy prior to chemotherapy, suggesting most patients had asymptomatic and/or small BM. In addition, it was not possible to evaluate intracranial responses in these patients. Therefore, we retrospectively evaluated the effects of ICIs plus chemotherapy in patients with symptomatic and asymptomatic BM. In addition, we examined both intracranial and extracranial responses to immunochemotherapy in patients who did or did not receive prior brain radiotherapy.
Patients and Methods
Patients with advanced NSCLC who received ICIs (pembrolizumab or atezolizumab) plus chemotherapy at Hiroshima City Hiroshima Citizens Hospital from January 2019 to July 2021 were retrospectively enrolled into this study. The characteristics and clinical data of patients prior to administration of ICIs plus chemotherapy were obtained. This study was approved by the Ethical Review Board of Hiroshima City Hiroshima Citizens Hospital (approval number No. 2019-62, July 11, 2019). Patient approval or the requirement for informed consent was waived because this study was performed during routine clinical practice.
Patients were divided into two groups according to presence of BM. The BM group included patients with BM before treatment with ICIs plus chemotherapy, and the Non-BM group included patients who did not have BM before treatment. In addition, the BM group was subdivided into two groups according to whether patients received radiotherapy to the brain. The brain metastasis-radiotherapy (BM-RT) group included patients who received brain radiotherapy before treatment with ICIs plus chemotherapy, and the BM-no RT group included patients who did not receive brain radiotherapy before treatment.
PFS was measured from the date of initiation of ICIs plus chemotherapy to the date of initial disease progression, death from any cause, or date last known to be alive without disease progression. OS was measured from the date of initiation of ICIs plus chemotherapy to death from any cause or date last known to be alive. The objective response rate (ORR) and disease control rate (DCR) were evaluated using the Response Evaluation Criteria for Solid Tumors, version 1.1 (11). Differences between groups were assessed by Mann-Whitney U-test and Fisher’s exact test. The Kaplan-Meier method was used to estimate PFS and OS rates, and the log-rank test was used to determine differences in survival rates. All statistical analyses were performed using EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria); EZR is a modified version of R commander designed to add frequently used statistical functions (12). p-Values <0.05 were considered statistically significant.
Results
Patient characteristics. Of 103 patients with advanced NSCLC, 26 patients were assigned to the BM group and 77 patients were assigned to the Non-BM group (Table I). No significant difference was observed in clinical status between patients in the two groups. PD-L1 expression was evaluated in 96 patients, and no difference in PD-L1 TPS was observed between groups. Driver mutations, such as epidermal growth factor receptor (EGFR) gene mutations, anaplastic lymphoma kinase (ALK) gene translocation, ROS proto-oncogene 1 (ROS-1) rearrangements, BRAF V600E mutation, MET exon 14 skipping mutation, and rearranged during transfection (RET) fusion, were observed in 14 patients. In the BM group, the median number of BMs was 2 (range=1-15), and the median maximum diameter of BMs was 15.5 mm (range=3.0-52.0 mm). Neurologic symptoms were observed at presentation in 8 patients. Radiotherapy for BM was performed in 16 patients, stereotactic radiosurgery (SRS) was performed in 12 patients, and whole-brain radiation therapy (WBRT) was performed in 4 patients (Table II).
Patient characteristics (N=103).
Characteristics in the BM groups (n=26).
Efficacy. No significant difference in ORR was observed between the BM and Non-BM groups (57.7% vs. 50.6%; p=0.651). However, the progressive disease (PD) rate was higher in the BM group than in the Non-BM group (30.8% vs. 11.7%). The intracranial ORR and DCR were 65.4% and 73.1%, respectively (Table III). Patients in the BM group tended to have worse PFS [median PFS, 5.3 months versus 8.6 months; hazard ratio (HR)=1.63; 95% confidence interval (CI)=0.92-2.90; p=0.092] and OS (median OS, 12.2 month versus 18.4 months; HR=1.97; 95%CI=0.99-3.88; p=0.050) compared with those in the Non-BM group (Figure 1A and B).
Best overall response and intracranial response.
Kaplan-Meier curves illustrating progression-free survival (A) and overall survival (B) in the brain metastasis (BM) and non-BM groups, respectively, and progression-free survival (C) and overall survival (D) in the brain metastasis-radiotherapy (BM-RT) and BM-no RT groups, respectively.
Radiotherapy for BM. Radiotherapy for BM was performed in 16 patients. No significant difference in age, sex, Eastern Cooperative Oncology Group performance status (ECOG PS), histology, stage, smoking status, driver mutation status, or PD-L1 status was observed between the BM-RT group and the BM-no RT group. However, the maximum size of BM and the proportion of patients with symptomatic BM were significantly larger in the BM-RT group (Table II). As shown in Table IV, the intracranial ORR and DCR were significantly higher in the BM-RT group than in the BM-no RT group (p=0.009 and 0.005, respectively). Regarding best overall response, ORR and DCR were significantly higher in the BM-RT group than those in the BM-no RT group (p=0.043 and 0.001, respectively) (Table IV). In addition, patients in the BM-RT group experienced longer PFS (median PFS, 6.4 months versus 3.2 months; HR=0.22; 95%CI=0.07-0.63; p=0.005) and OS (median OS, 17.8 months versus 5.9 months; HR=0.26; 95%CI=0.08-0.82; p=0.022) compared with patients in the BM-no RT group (Figure 1C and D). These results suggest that patients who received prior brain radiotherapy experienced a superior therapeutic benefit of ICIs plus chemotherapy despite having larger and more symptomatic BM. Notably, even in patients with low PD-L1 expression (TPS <50%), extracranial disease control was obtained in 5 of 6 patients (83.3%) in the BM-RT group, whereas it was obtained in only 1 of 7 patients (14.3%) in the BM-no RT group (Table V).
Best overall response and intracranial response in patients with and without brain metastasis (BM)-radiotherapy (RT).
Characteristics and response of 26 patients with brain metastasis (BM) based on prior radiotherapy (RT) status.
Safety. Adverse events (AEs) in the BM-RT and BM-no RT groups are listed in Table VI. The most frequent grade 3 or higher AE was neutropenia in both groups. Two patients in the BM-RT group developed radiation necrosis of BM after brain radiotherapy, but treatment of radiation necrosis was not performed because only imaging findings were available. Overall, 3 patients (18.8%) in the BM-RT group and 3 patients (30.0%) in the BM-no RT group experienced treatment-related AEs leading to discontinuation of chemoimmunotherapy, and there were no-treatment related deaths.
Adverse events in 25 patients with brain metastasis (BM) based on prior radiotherapy (RT) status.
Discussion
This study demonstrated the efficacy of ICIs plus chemotherapy for BM in NSCLC patients who did or did not receive prior brain radiotherapy. Patients with BM tended to have shorter PFS and OS compared with those without BM. However, patients who received brain radiotherapy before chemoimmunotherapy had better PFS and OS than those who did not receive brain radiotherapy. Notably, intracranial and extracranial ORRs were similarly high in these patients, despite having low PD-L1 expression.
To our knowledge, this is the first study to demonstrate an association between the efficacy of ICIs plus chemotherapy and prior brain radiotherapy in NSCLC patients with BM. Although a pooled analysis validated the efficacy of ICIs plus chemotherapy in patients with BM, only a small proportion of patients with symptomatic BM were included (10). In addition, Afzal et al. reported the efficacy of chemoimmunotherapy in NSCLC patients with symptomatic BM in clinical practice (13). However, this study included only a small number of patients, and the effects of prior brain radiotherapy on chemoimmunotherapy efficacy were unclear. In the present study, patients who received brain radiotherapy before chemoimmunotherapy had a better prognosis than those who did not. Even in patients with symptomatic and large BM, an improved long-term prognosis may be expected following treatment with a combination of prior brain radiotherapy and ICIs plus chemotherapy.
Several reports have demonstrated the efficacy of immunotherapy combined with brain radiotherapy in patients with BM from NSCLC (14, 15). The underlying mechanism for the better outcomes associated with immunotherapy plus brain radiotherapy is proposed as follows. Radiation promotes the release of tumor neoantigens, which activate and stimulate proliferation of CD8+ tumor-specific T cells (16). In addition, brain radiotherapy may facilitate T-cell permeability by disrupting the blood-brain barrier (17). Moreover, ICIs are more effective in patients with high CD8+ tumor infiltrating lymphocytes, as well as PD-L1 expression in tumor tissues (18). These mechanisms have the potential to contribute to synergistic effects of brain radiation and immunotherapy against BM.
The present results indicate that intracranial and extracranial ORRs were similarly high in patients who received brain radiotherapy. Although the mechanism underlying the dual responses of brain radiotherapy before chemoimmunotherapy remains unclear, an extracranial abscopal effect induced by brain radiotherapy may be involved. An abscopal effect can be broadly defined as a reaction outside an irradiated area but within the same organism (19). Although the central nervous system had been considered to have immune privilege due to the blood-brain barrier, several cases of abscopal effects induced by localized brain radiotherapy have been reported. Grimaldi et al. reported that 7 patients with melanoma treated with ICIs showed abscopal effects followed by brain radiotherapy (20). In addition, one case of an extracranial abscopal effect induced by combination immunotherapy and brain radiotherapy in a patient with lung cancer has been reported (21). Therefore, extracranial abscopal effects may have been induced by the combination of brain radiotherapy and immunochemotherapy in the present cases.
In the present study, the intracranial and extracranial ORRs were similarly low in patients treated with chemoimmunotherapy alone, particularly in patients with low or no PD-L1 expression. In contrast, the ORRs were similarly high in patients treated with a combination of prior brain radiotherapy and chemoimmunotherapy, even among those with low or no PD-L1 expression. The better outcomes of combination treatment in patients with low or no PD-L1 expression may be due to enhancement of PD-L1 expression by radiation. Indeed, radiation has been reported to promote PD-L1 expression in cancer cells through oncogenic signaling, such as through the phosphatidylinositol-3-kinase (PI3K)/AKT signaling pathway (22). Furthermore, in a case report of a patient with NSCLC with BM, PD-L1 expression in the BM increased continuously after repeated brain radiotherapy, indicating that radiotherapy may also promote PD-L1 expression in metastatic brain tumor cells (23). These reports suggest that a combination of prior brain radiotherapy and chemoimmunotherapy may be beneficial, particularly in patients with low or no PD-L1 expression.
Treatment with ICIs and brain radiotherapy has been reported to not be associated with a significant increase in radiotherapy-related adverse events (24). In the present case, two patients developed radiation necrosis of BM following brain radiotherapy, although they were asymptomatic. Regarding necrosis of BM, Martin et al. (25) observed an association between administration of immunotherapy and symptomatic radiation necrosis in patients undergoing brain radiotherapy. However, the risk of necrosis was strongly associated with the use of ipilimumab for melanoma, and no significant association was found between brain necrosis and PD-1/PD-L1 inhibitors combined with brain radiotherapy for NSCLC. Prospective studies are needed to better characterize the risks and benefits of combining brain radiotherapy with immunotherapy in patients with NSCLC with BM.
The present study has several limitations. First, this was a retrospective, single-center study with a small sample size. However, few reports verifying the efficacy of chemoimmunotherapy and brain radiotherapy before chemoimmunotherapy in patients with NSCLC with symptomatic BM have been published. Therefore, we believe that our study is important as it contributes real-world evidence for clinical practice. Second, most patients with asymptomatic and/or small BM were not treated with brain radiotherapy before chemoimmunotherapy. It is unclear whether prior brain radiotherapy will be as effective for these patients as for those with symptomatic and large BM. It is necessary to carefully consider whether prior brain radiotherapy should be performed for these patients, because brain radiotherapy is associated with side effects, such as brain necrosis.
Footnotes
Authors’ Contributions
Y.T. and H.S. devised the project and main conceptual ideas. Y.T. and J.Y. collected the clinical data of the patients and performed the retrospective chart review. R.S. and S.M. supervised the project. All Authors provided final approval for publication.
Conflicts of Interest
The Authors declare no conflicts of interest associated with this study.
- Received July 20, 2022.
- Revision received August 8, 2022.
- Accepted August 9, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
