We enrolled patients with resectable stage IIB-IIIB NSCLC (according to the staging criteria of the American Joint Committee on Cancer, 8th edition) without previous anticancer therapy, an Eastern Cooperative Oncology Group performance-status score of 0 or 1 (on a 5-point scale in which higher scores reflect greater disability) without EGFR/ALK mutation. Patients had to have measurable disease according to the Response Evaluation Criteria in Solid Tumors, version 1.1, and pretreatment tumor tissue available to assess the expression of programmed death ligand 1 (PD-L1). Patients with known ALK translocations or EGFR mutations were excluded. The inclusion criteria were: 1) cytological or histological confirmation of NSCLC; 2) primary staged as clinical or pathological stage IIB-IIIB; 3) completed at least one cycle of immunotherapy. The exclusion criteria were 1) EGFR/ALK mutation, 2) important organ dysfunction before treatment; 3) did not complete a follow-up visit of the first cycle at the time of data collection; 4) previously accepted anti-tumor treatment.

In the prospective study, NSCLC patients were treated with neoadjuvant Toripalimab (240mg, every 3 weeks) plus double platinum-based chemotherapy from December 2020 to March 2022 at Peking University Cancer Hospital. Patients received cisplatin 75mg/m2 and pemetrexed 500 mg/m2 for adenocarcinoma, or cisplatin 75mg/m2 and nab-paclitaxel 260 mg/m2 for other subtypes on day 1-2 of each 21-day cycle. And intravenous Toripalimab was used on day 1 with chemotherapy of each 21-day cycle. Surgical resection was performed 6-8 weeks afterward. We analyzed the data of patients in the Renaissance study and explored the association between clinical outcomes and irAEs after the commencement of neoadjuvant Toripalimab treatment. The study was approved by the Peking University Ethic board (2020YJZ58) and registered at https://clinicaltrials.gov/ as NCT04606303. All patients signed written informed consent forms, and all data were deidentified.

All patients were staged according to the 8th Edition American Joint Committee on Cancer staging system. Characteristics of patients were summarized, including age, sex, pathological type, smoking history, tumor-node-metastasis (TNM) stage, immunotherapy-related adverse events (irAEs), and residual viable tumor (RVT).

The primary endpoints were major pathological response (MPR) rate (19) (≤10% residual viable tumor cells in the primary tumor and sampled lymph nodes). The second end points were the emergence of irAEs, pathological complete response (pCR) rate (0% residual viable tumor cells in the primary tumor and sampled lymph nodes), objective response rate, R0 resection rate, perioperative safety, and event-free survival.

IrAEs were defined as having a potential immunological basis that required more frequent monitoring and potential intervention assessed by a multidisciplinary team of physicians, surgeons, and radiologists. IrAEs were classified according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.03 (20) and were handled according to the guidelines on the management of immunotherapy-related toxicities (21, 22). The time to onset of irAE was defined as the time from the start of immunotherapy to the occurrence of irAE. Event-free survival (EFS) was defined as the time from the start of immunotherapy to any progression of disease precluding surgery, progression or recurrence of disease after surgery, progression of disease in the absence of surgery, or death from any cause.

Descriptive statistics of baseline, clinicopathological, and operation-related characteristics were listed. The pCR, MPR, R0 resection rate, the occurrence of irAEs, the time interval between Toripa, limab use and the occurrence of irAE, and event-free survival were calculated. SPSS 26.0 (IBM, Armonk, NY, USA) was used for statistical analysis. The continuous data were presented as medians (ranges) and analyzed using the Mann-Whitney U-test. The categorical data were presented as numbers (percentages) and analyzed using the chi-square or Fisher’s exact test. And the survival outcome data were presented as average (ranges) and analyzed by the Cox proportion risk regression model. The Kaplan-Meier curve was analyzed by Log-Rank analysis. Reported P values are two-sided, and the significance level was set at 0.05 for all analyses unless otherwise noted.

From December 2020 to March 2022, 55 patients met the inclusion criteria, the median age was 62 years old (IQR 55-66 years old) ( Figure 2 ). Overall, 50 patients (90.9%) were men, 49 patients (89.1%) had a smoking history, 16 patients (29.1%) were stage IIB, 31 patients (56.4%) were stage IIIA, 8 patients (14.5%) were stage IIIB. Nine patients (16.4%) were staged as N0, 16 patients (29.1%) were staged as N1, and 30 patients (54.5%) were staged as N2. Forty-four patients were diagnosed with squamous cell carcinoma (80%), nine patients were diagnosed with adenocarcinoma (16.4%), and 2 patients were proven to have adenosquamous lung cancer.

About 69.1% of patients accepted 2 cycles of neoadjuvant Toripalimab plus chemotherapy, 14 patients accepted 3 cycles of neoadjuvant Toripalimab plus chemotherapy (25.5%), one patient received 4 cycles of neoadjuvant Toripalimab plus chemotherapy (1.8%), and 2 patient received one cycle of neoadjuvant Toripalimab plus chemotherapy (3.6%) ( Table 1 ) One of the 2 patients that only take 1 cycle was due to bile tract infection, another patient suffered severe immune-related enteritis and can’t continue immunotherapy. TRAE occurred in 53 patients (96.4%), which contains myelosuppression, vomiting, nausea, alopecia, acroanesthesia, anorexia, hepatic function impairment, etc. Among them, 17 patients (30.9%) suffered Grade 3-4 TRAE. While there were 22 patients (40%) involved in this clinical trial were diagnosed with irAEs. Immune-related thyroid dysfunction, rash/pruritus, enteritis, and pneumonitis occurred in 12 patients (21.8%), 7 patients (12.7%), 2 patients (3.6%), and 1 patient (1.8%), respectively. Among these 22 patients, 6 patients suffered Grade 2-3 irAEs. The common Grade 2-3 irAEs were rash and pruritus (9.1%), immune-related enteritis (9.1%), immune-related pneumonitis (4.5%), and thyroid dysfunction (4.5%) ( Table 2 ). The median time between the commencement of Torapalimab and diagnosis of irAEs was 28.5 days (IQR: 18.3-77.5 days). And the median time interval between the commencement of Torapalimab and diagnosis of grade 1, grade 2, and grade 3 irAEs were 32.5 days,21 days, and 62 days. Grade 1-2 irAEs appear to occur earlier than Grade 3 ( Figure 3 ). Grade 1-2 rash was usually treated withsteroids ointment and oral antihistamines. Grade 1-2 thyroid dysfunction was treated with thyroid hormone supplements. While immune-related pneumonitis and enteritis were treated with intravenous steroids injection and sequentially dose decreased till suspension, symptomatic treatment was also used when patients suffered serious dyspnea or diarrhea. In our research, two patients with grade 2 irAEs (9.1%) and 3 patients with grade 3 irAEs (13.6%) were treated with glucocorticoid therapy. The median glucocorticoid-using time duration was 12 days (IQR: 7-79 days).

Forty-eight of 55 patients finally received thoracic surgery with a median time interval of 67 days (IQR 39-113 days) after neoadjuvant treatment. Pathology confirmed that 31 (64.6%) patients achieved MPR and 24 (50.0%) patients achieved pCR.

In a total of 48 patients who received R0 resection, eleven patients suffered from thyroid dysfunction, of which six patients achieved MPR and 5 patients achieved pCR, the median time from treatment initiation to onset was 45 days (IQR 21-91 days). Six patients with rash or pruritus achieved MPR, and 5 patients achieved pCR, the median time to onset was 8 days (IQR 6-29 days). One patient with enteritis achieved pCR, and the time interval between treatment and the onset of enteritis was 16 days. Among these patients with irAEs, three patients suffered grade 2-3 irAEs and achieved pCR( Table 3 ).

Furthermore, as to the safety of thoracic surgery after neoadjuvant Toripalimab plus chemotherapy, irAEs showed no significant influence on the operation time (170.6 min vs 165.7 min, p=0.775) ( Figure 1A ), intraoperative blood loss (67.4 mL vs 64.3 mL, p=0.831) ( Figure 1B ) and preoperative waiting time (93 days vs 97 days, p=0.630) ( Figure 1C ). The pathological response outcomes of different stage of NSCLC was presented in Table 4 . It showed that Stage IIB to IIIA might benefit mostly from neoadjuvant immunochemotherapy, and no significant difference was found between clinical stage and pathological response (p=0.27).

We also analyzed the time-onset of irAE and the time interval between the last cycle of neoadjuvant therapy and surgery ( Figure 4 ), the p-value didn’t show a significant difference, which further confirms that the occurrence of irAE would not delay surgery and neoadjuvant immuno-chemotherapy could be estimated as dependable and safe.

There is one patient who decided to watch and wait after complete clinical response (CCR) ( Figure 5 ). This patient was diagnosed with lung adenocarcinoma, cT2N2M0, stage IIIA. His CT examination in May 2021 revealed truncated bronchial stenosis in the upper lobe of the right lung, irregular soft tissue foci next to the hilum, about 32*22mm, with distal lung tissue atelectasis. There were also multiple enlarged lymph nodes in the mediastinum and the right hilum. Bronchoscopy showed a bulge around 2/3 circumference of the right main bronchus, with the upper edge about 5mm from the bulge. Pathological biopsy suggested moderately differentiated squamous carcinoma. He was treated with 1 cycle of ABP (albumin- bounded paclitaxel) + cisplatin + Toripalimab in 2021/6/25. The next cycle of treatment was canceled due to the development of severe immune-related enteritis which was treated at Peking Union Medical College Hospital subsequently. On 2021-9-22 his chest CT showed: the soft tissue masses near the right hilum disappeared. Multiple enlarged lymph nodes were reduced in size. PET-CT of the right pulmonary hilum did not show hypermetabolism. In 2021-12-16, he reviewed bronchoscopy and the biopsy reported mild chronic inflammation of the mucosa of the pseudostratified ciliated columnar epithelium, with two heterogeneous cells visible in the interstitium, considered to be cancer. In 2022-4-13, we reviewed the bronchoscopy again, and the biopsy showed no definite clue of lung cancer. In 2022-12-16, we retested the chest CT: it showed that the soft tissue thickening in upper lobe of the right lung near the hilum was as before, either was the bronchial stenosis in the right upper lobe.

There were also 6 other patients who failed to receive surgery, of which three patients failed to receive surgery for the poor performance in radiological imaging, as well as 2 patients’ surgery were canceled respectively for the anatomical adhesion during surgical exploration and poor physical performance. Notably, one patient’s surgery was canceled for grade 3 immuno-related pneumonitis. This patient has diagnosed with stage IIIB adenocarcinoma without ALK translocation and EGFR mutation in the left lower lobe ( Figure 6A ), thus three cycles of neoadjuvant Toripalimab combined with chemotherapy were used for the surgical opportunity. While reduced exercise tolerance come out after the third cycle, he simultaneously had intermittent fever during hospitalization, chest computed tomography showed large-area ground-glass opacity with consolidation ( Figure 6B ), which confirmed the diagnosis of immuno-related pneumonitis. Arterial blood gas analysis showed an arterial partial pressure of oxygen of 68.2 mmHg. After standard treatment of glucocorticoid therapy combined with continued antibiotic treatment, the patient recovered from the server pneumonitis with continuous oxygen therapy. The patients achieved great improvement for 18 months from the treatment start ( Figure 6C ). And recently computed tomography showed that the tumor completely disappeared.

There are 14 patients who took adjuvant therapy, which includes chemo-immunotherapy with/without sequential immunotherapy for maintenance treatment, or directly adjuvant immunotherapy. There are 34 patients who didn’t take adjuvant therapy, the result of adjuvant therapy were presented in Table 5 .

The average EFS of patients without irAE (33 patients) was 22.7 months (95%CI 21.3-24.2 months). The median EFS of patients with irAE (22 patients) was 23.5 months (95%CI 21.0-25.9 months), and both median times were not reached, The log-rank analysis showed that the p-value was 0.61.

6 patients arrived at the end point of our study, 3 of them had encountered local recurrence, 2 of them had been diagnosed with metastatic lung cancer, and only 1 patient died during our study.

We separated the patients with irAE into the early-onset and late-onset, each separately included 14 and 8 patients. The watershed between them was 3 months. Due to the small sample size, the HR of early-onset irAE had no significant difference, and the HR of late-onset irAE was 4.53 (p=0.065, 95%CI 0.91-22.5). While the log-rank analysis of the K-M curve showed a significant difference (p=0.021) ( Figure 7 ).