Neoadjuvant immunotherapy for patients with non-small cell lung cancer-current evidence.

Shu and colleagues (1) made a significant contribution to a growing body of evidence for the utilization of immunotherapy in the neoadjuvant setting for patients with resectable non-small cell lung cancer () (2-4). Of 39 assessed patients, 31 were commenced on treatment consisting of Atezolizumab with nab-paclitaxel and carboplatin. After excluding two patients with colorectal cancer and brain metastasis, 29 underwent surgery with a curative intent. At the time of operation, three patients were deemed unresectable, with 26 patients achieving R0 resection. There was one (3%) mortality from pneumonia and respiratory failure, with major perioperative complications mostly related to neutropenia, liver dysfunction, and thrombocytopenia. Pathological analysis of resected specimens reported that 57% of the intention-to-treat population had major pathological response (MPR), including 33% who had pathological complete response (PCR).

Table 1

Study characteristics of neoadjuvant immunotherapy for patients with resectable non-small cell lung cancer

Author	Institution	Patients, n, %*	F/U (months)	Neoadjuvant immunotherapy	Neoadjuvant chemotherapy	Pathological complete response primary site	Pathological complete response primary site and LNs	Major pathological response	Mortality
Shu 2020 (1)	Columbia University; MGH; Dana Farber Cancer Institute	29/30 (97%)	13	Atezolizumab 1,200 mg; day 1 of 4 q3w cycles	Paclitaxel 100 mg/m2 on days 1, 8, 15 + Carboplatin, AUC =5 day 1 of 21d cycle	10/30 (33.3%)^	NS	17/30 (56.7%)^	1/29 (3.4%)
Gao 2020 (2)	PUMC	37/40 (93%)	3	Sintilimab 200 mg 2 cycles q3w; 4 to 6 weeks preoperative	None	6/37 (16.2%)	3/37 (8.1%)	15/37 (40.5%)	2/37 (5.4%)
Bott 2019 (3)	Johns Hopkins; MSKCC	20/22 (91%)	20	Nivolumab; 3 mg/kg; 4 and 2 weeks preoperative	None	3/20 (15%)	2/20 (10%)	9/20 (45%)	0/20 (0%)
Yang 2018 (4)	Duke University	13/24 (54%)	24	Ipilimumab, 10 mg/kg; day 1 of neoadjuvant cycles 2, 3	3 cycles of Paclitaxel, 175 mg/m2 + Cisplatin, 75 mg/m2 or Carboplatin, AUC =6	2/13 (15.4%)	NS	NS	0/13 (0%)

*, number of patients who underwent resection out of patients treated with neoadjuvant immunotherapy; ^, intention-to-treat population. MGH, Massachusetts General Hospital; PUMC, Peking University Medical Centre; MSKCC, Memorial Sloan Kettering Cancer Center; F/U, follow-up; LN, lymph nodes; AUC, area under curve; NS, not specified.

Compared to previous studies, a significant proportion of patients (10.3%) had exploratory ‘open-and-close’ operations that concluded the tumor was unresectable (2-4). This may reflect a relatively higher proportion of patients (77%) with advanced cIIIA disease, and willingness of the surgeons to attempt challenging resections for larger tumors with mediastinal nodal involvement. Notably, 54% of patients underwent thoracotomy rather than minimally-invasive surgery, reflective of the technical challenges posed by infiltration of inflammatory cells associated with the pseudoprogression phenomenon related to immunotherapy. Previous studies described frequent conversions to open thoracotomies after encountering hostile hilar fibrosis and adhesions (3). In regard to patient selection, non-smokers were excluded from the study, and a relatively higher proportion of patients (40%) with squamous cell carcinoma (SCC) were included, compared to previous studies (3,4). Of the ten patients with SCC who underwent R0 surgical resection, 80% had MPR and 50% had PCR, compared to 53% and 33% of patients who had adenocarcinomas, respectively. This echoed findings by Gao et al., who analyzed neoadjuvant Sintilimab (2). In their report, out of 37 patients who had evaluable pathology, 15 (41%) had MPR, all of whom had SCC, with a strong trend favoring response for SCC compared to adenocarcinomas. Although limited in numbers, these findings support the hypothesis that neoadjuvant immunotherapy may be more effective in patients with malignancies that have a high mutational burden, which may be associated with SCC and smoking history (3,5).

Overall, this study further demonstrated that neoadjuvant immunotherapy was well tolerated prior to surgical resection with an acceptable safety profile. Larger studies in the future should assess the impact of PD-L1 expression and correlation of pathological response with radiological response, which remains uncertain (2,3).

The article’s supplementary files as