Extensive‐stage small cell lung cancer: Is prophylactic cranial irradiation necessary in the era of immunotherapy with MRI surveillance?

The role of prophylactic cranial irradiation (PCI) in treating extensive‐stage small‐cell lung cancer (ES‐SCLC) has been controversial. This study aimed to comprehensively analyze the efficacy of PCI for the treatment of ES‐SCLC under active brain magnetic resonance imaging (MRI) surveillance.


INTRODUCTION
Small-cell lung cancer (SCLC) is an aggressive neuroendocrine cancer that grows rapidly and has a high risk of early metastasis. Approximately 15-33% of patients with SCLC have brain metastases (BM) upon initial diagnosis, and 80% of patients develop BM within 2 years. [1][2][3][4] Prophylactic cranial irradiation (PCI) is the standard method used for treating patients with limited-stage SCLC that have shown a complete response to chemoradiotherapy. 5,6 However, the value of PCI in patients with extensive-stage SCLC (ES-SCLC) is in debate. 7,8 The randomized trial initiated by the European Organization for

Research and Treatment of Cancer Radiation Oncology Group and
Lung Cancer Group reported that PCI prolonged ES-SCLC patient survival and reduced the incidence of BM. 7 With opposing conclusions, the Japanese randomized trial demonstrated no reduction in overall survival (OS) when PCI was replaced by active magnetic resonance imaging (MRI) surveillance. 8 More recently, two randomized phase III trials (IMpower133 and CASPIAN) indicated that anti-programmed cell death ligand 1(PD-L1) therapy (atezolizumab or durvalumab) significantly improved OS with a tolerable safety profile in patients with ES-SCLC. [9][10][11] Based on the outstanding findings, anti-PD-L1 immunotherapy plus chemotherapy has become the new first-line standard-of-care for the treatment of ES-SCLC. However, PCI was not mandated in the IMpower 133 trial, which was administered to only over 11% of patients enrolled, and no rates of BM or neurological death were reported. PCI is also poorly controlled in CASPAIN, is only available to patients undergoing chemotherapy alone, and is only delivered to 8% of the patients. There are limitations to our understanding of the benefits of introducing immunotherapy to PCI because of the lack of proper controls in the aforementioned randomized trials.
In this era of immunotherapy, no prospective studies have demonstrated the benefits of PCI in patients with ES-SCLC. In view of these concerns, we performed this multi-institutional real-world study to investigate the efficacy of PCI for patients with ES-SCLC who responded well to initial chemoimmunotherapy in the presence of active brain MRI surveillance.

Treatment and follow up
All patients received at least four cycles of platinum plus etoposide During the maintenance phase, patients in the PCI group received cranial radiation at a total dose of 25 Gy delivered in 10 daily fractions (2.5 Gy per fraction) using intensity-modulated radiotherapy with hippocampal avoidance. PCI was initiated within 8 weeks of chemotherapy. MRI monitoring of the intracranial status was conducted at least once every 2-3 months in all patients, regardless of whether PCI was performed. Consolidative thoracic radiotherapy for patients with a good response to the initial chemoimmunotherapy and palliative radiotherapy for sites other than the thorax for symptom alleviation were permitted for both groups.
Treatment for intracranial relapse and other post-progression treatments in the two groups were left to the treating physician's discretion.

Statistical analysis
The

Overall survival analysis
The Kaplan-Meier survival curves for OS are shown in Figure 1A.

Progression-free survival analysis
At the data cutoff, 13 Figure 1B). At 6 and 12 months, the PFS rates in the PCI group were 76.50% and 40.30%, respectively, while those in the non-PCI group were 70.70% and 25.70%, respectively. The univariate and multivariate models of PFS predictors are shown in Table 3. Univariate Cox regression analysis showed that none of the factors were significant prognostic indicators of PFS. Multivariate analysis showed that the factors associated with increased PFS were the absence of smoking and consolidative thoracic radiotherapy.

Brain metastasis
Compared to the non-PCI group, the cumulative incidence rate of BM was lower in the PCI group (12 and 18 months: 12.79% and 27.48% vs. 38.09% and 40.22%, respectively) when counting death as a competing risk, but the difference was not statistically significant (Gray's test, p = 0.14; Figure 2). Information about BM initial salvage treatment was as follows: salvage radiotherapy for BM was administered to one (25.00%) of four patients with BM in the PCI group and nine (45.00%) of 20 patients with BM in the non-PCI group.

DISCUSSION
This first real-world study aimed to analyze the feasibility of using PCI under active brain MRI surveillance for ES-SCLC in the immunotherapy era and found that neither improvement in OS or PFS nor a decrease in the risk of developing BM was associated with PCI.
Notably, the median OS and PFS seen in this study were little longer than those observed in previous IMpower 133 ( However, these two drugs were not included in our study because serplulimab and adebrelimab plus chemotherapy had not been approved for patients with ES-SCLC prior to the initiation of this study.
In contrast to two randomized studies in Europe and Japan, PCI did not achieve a statistically significant difference in reducing BM in this study. 7,8 Immune checkpoint inhibitors that can penetrate the blood-brain barrier may account for this observation. Atezolizumab prolonged the time to intracranial progression according to a subgroup analysis of the IMpower133 study. 15 Moreover, durvalumab decreased the risk of central nervous system relapse in the Pacific Phase III NSCLC study. 16 Therefore, PCI may be replaced by immune checkpoint inhibitors in the treatment of microscopic central nervous system diseases. 17 Increased survival has led to greater concerns regarding long-term nerve injury following PCI. Many studies have reported high frequencies of cognitive dysfunction and memory impairment after wholebrain radiotherapy. [18][19][20] MRI surveillance and salvage radiotherapy may have greater benefits for the cognitive function and quality of life than early radiation of the entire brain. Physicians should evaluate each patient's specific situation before providing comprehensive treatment to improve the survival and quality of life.
This study was limited by its retrospective nature, which means that selection bias and heterogeneity were inevitable among the enrolled patients. Second, given the small sample size and balanced baseline characteristics, propensity score matching was not performed to equalize the number of patients in the two groups and minimize selection bias. Third, as the toxicity of PCI and that associated with chest radiotherapy may be confounded, no toxicity-related data were reported in this study.

CONCLUSIONS
The increasing use of brain MRI surveillance and immunotherapy for ES-SCLC has further diminished the potential use of PCI in this population. Further studies are warranted to assess the efficacy and safety of PCI in patients with ES-SCLC receiving immunotherapy.

ACKNOWLEDGMENTS
We thank all the patients, their families, and investigators involved in this study.

CONFLICT OF INTERESTS STATEMENT
There are no conflict of interests.

ETHICS APPROVAL AND INFORMED CONSENT
All procedures involving human participants performed in this study were in accordance with the ethical standards of the local ethics committees (XYFY2022-KL355-01) and the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.