Bauman et al. [7] reported that the most common reasons for thoracic salvage surgery were obvious relapse as shown by computed tomography, persistently abnormal 18F-fluorodeoxyglucose positron emission tomography findings after completion of radiotherapy, and a delayed decision to convert to a trimodal approach. However, thoracic salvage surgery can be defined as surgical resection of residual or recurrent primary lung tumors after previous local or systemic treatments [8]. Several retrospective studies have shown the safety and feasibility of salvage surgery after initial treatment with TKIs or ICIs for advanced NSCLC [4, 5, 9]. Interestingly, Nakanishi et al. [10] reported effective local control and favorable survival outcomes by salvage surgery for patients with stage IIB to IIIB small cell lung cancer who received cisplatin, etoposide, and radiotherapy. Recent clinical trials of neoadjuvant treatment and adjuvant treatment for resectable NSCLC have shown that epidermal growth factor receptor-TKI and ICI therapy might provide a superior DFS benefit [11–14]. In the CheckMate 816 trial, Forde et al. [11] showed that neoadjuvant nivolumab plus chemotherapy resulted in significantly longer event-free survival and a higher percentage of patients with a pCR by chemotherapy alone among patients with stage IB to IIIA resectable NSCLC. In a phase-III trial (ADAURA trial) of patients with stage IB to IIIA epidermal growth factor receptor mutation-positive NSCLC, Wu et al. [12] found that DFS was significantly longer among those who received osimertinib than among those who received placebo. Additionally, the IMpower010 trial showed a DFS benefit with atezolizumab versus best supportive care after adjuvant chemotherapy in patients with resected stage II to IIIA NSCLC, with a pronounced benefit in patients whose tumors expressed programmed death ligand-1 (PD-L1) on ≥ 1% of tumor cells [13]. Furthermore, O’Brian et al. [14] performed an interim analysis of a randomized phase-III trial (KEYNOTE-091) and found that pembrolizumab significantly improved DFS compared with placebo and was not associated with new safety signals in completely resected, PD-L1-unselected, stage IB to IIIA NSCLC. These results might affect and improve the survival of patients with advanced NSCLC and even those with small cell lung cancer [10]. In the present study, the initial treatment regimen varied, including cytotoxic chemotherapy with an ICI, a single ICI, and double ICIs. The study also included patients during transition from separate treatment to combination treatment of cytotoxic agents and ICIs. We basically selected pembrolizumab as immunotherapy if the patient’s TPS was ≥ 50% and either nivolumab ± ipilimumab or atezolizumab if the TPS was < 50%. Essentially, we now choose combination therapy using both cytotoxic agents and ICIs except in patients aged > 80 years. These indications might differ among institutions.
Salvage surgery is generally considered technically more difficult with potentially higher morbidity than alternative treatments. Therefore, careful patient selection combined with surgical expertise can allow for successful salvage surgery with minimal morbidity [8]. Our data showed that OS was significantly better in patients with ypStage 0 to I cancer than in those with ypStage ≥ II cancer. Salvage surgery provides patients who were initially diagnosed with unresectable advanced NSCLC a chance of long-term survival, especially those who have achieved downstaging to ypStage 0 to I. These patients are probably good candidates for salvage surgery. Although we were able to safely perform salvage surgery, our study showed that the rate of grade ≥ 3 perioperative complications reached 23.1% and included respiratory failure and gastrointestinal hemorrhage. We also experienced one patient with a grade 5 complication (gastrointestinal hemorrhage), resulting in 90-day mortality. These severe perioperative complications might have affected the patients’ prognosis, and such patients might have lived longer if they had undergone continuation of chemotherapy rather than salvage surgery. In addition to these perioperative complications, the patients also developed many adverse events due to chemotherapy. Therefore, the candidates for salvage surgery should be selected with caution. However, we cannot estimate preoperatively which factors might contribute to perioperative severe complications. Therefore, we must carefully evaluate patients’ clinicopathological factors to select good candidates for salvage surgery in a multi-institutional prospective study. If severe postoperative complications related to the surgical procedure and immunotherapy occur, they should be managed as early as possible and the patients should be provided more intensive care.
We continued ICI treatment after salvage surgery except in patients with a pCR; however, the optimal time point at which to stop the ICI treatment was unknown. Recent reports including the IMpower 010 trial [13] and KEYNOTE-091 trial [14] showed that ICI treatment may be continued for around 1 year postoperatively as adjuvant chemotherapy. According to these reports, we could have stopped the ICI treatment at around 1 year. However, previous trials included only patients with pStage II to IIIA cancer. Our study included patients with more advanced cancer ranging from cStage IIIA to IVB. The difference in these populations might have affected the prognosis, and further prospective studies are therefore needed. Another serious issue that needs to be resolved is the lack of reliable biomarkers for predicting the effect and prognosis of immunotherapy. Several promising biomarkers are currently available, including PD-L1 expression, the tumor mutation burden, and gene expression signatures [15]. However, these factors are not reliable. After reviewing the clinical value of the NLR in patients with NSCLC treated with ICIs, Jiang et al. [16] set the NLR cutoff value at 5 and concluded that a high blood NLR (> 5) is associated with shorter progression-free survival and OS in patients treated with ICIs; they therefore suggested that the NLR has potential predictive and prognostic value. We also previously reported the usefulness of the NLR in a case report in which the patient, who was included in the present study, achieved a pCR after immunotherapy [17]. In the present study, the NLR of each patient decreased after immunotherapy and persistently remained between 1 to 2 during immunotherapy (Fig. 1). The NLR is simple to measure and might serve as a reliable biomarker of immunotherapy.
Our study has some limitations, including its retrospective nature, small number of patients, and heterogeneous population (e.g., cStage IIIA–IVB, various histologic types, and various immunotherapy regimens). A prospective study is needed to evaluate the safety and feasibility of salvage surgery for patients with initially unresectable lung cancer. We should also compare the prognosis of patients with advanced lung cancer who undergo salvage surgery versus those who undergo multidisciplinary treatment without salvage surgery. Such studies will show the real efficacy of salvage surgery.
Local treatment such as salvage surgery in addition to systemic chemotherapy has an important role in the control of advanced lung cancer. Prospective studies of salvage surgery after ICI therapy are urgently needed to improve the prognosis of initially unresectable advanced NSCLC.