A multicenter, single-arm, open study of neoadjuvant or conversion atezolizumab in combination with chemotherapy in resectable small cell lung cancer (Cohort Study)

Abstract

Background:

This study aimed to investigate the prospects of using chemotherapy in combination with atezolizumab in the neoadjuvant or conversion treatment of small cell lung cancer (SCLC).

Methods:

Prior to surgery, untreated patients with limited-stage SCLC received three cycles of neoadjuvant or conversion atezolizumab combined with chemotherapy of etoposide and platinum. The primary endpoint of the trial was pathological complete response (pCR) in the per-protocol (PP) cohort. In addition, safety was assessed based on treatment-related adverse events (AEs) and postoperative complications.

Results:

Overall, 13 of 17 patients (including 14 males and 3 females) underwent surgery. In the PP cohort, pCR and major pathological response were observed in 8 (8/13, 61.5%) and 12 (12/13, 92.3%) patients, respectively. According to the intention-to-treat (ITT) analysis, the pCR and major pathological response in the ITT cohort were 47.1% (8/17) and 70.6% (12/17), respectively. In addition, an overall response rate of 100% was recorded in the PP cohort. Moreover, 15 (15/17, 88.2%) patients and 1 (1/17, 5.9%) in the ITT cohort attained partial remission (PR), and complete remission, respectively, with an overall response rate of 94.1%. The median overall survival of the patients of pCR and the median event-free survival of the patients on surgery had not achieved. However, the median overall survival of the patients of non-pCR was 18.2 months and the median event-free survival of the nonsurgical patients was 9.5 months. During the neoadjuvant treatment, the incidence of grade 3 or higher AEs was 58.8% (10/17). Additionally, three patients (17.6%) developed immune-related adverse event (grades 1–2).

Conclusion:

In patients with SCLC, neoadjuvant or conversion atezolizumab combined with chemotherapy significantly improved pCR with manageable AEs. Therefore, this regimen may be considered a safe and effective treatment for SCLC.

Introduction

Highlights

• The NeoSCI trial was to investigate the efficacy and safety of neoadjuvant chemoimmunotherapy for small cell lung cancer.

• Surgery patients achieved a pathological complete response of 61.5%.

• Major pathological response of 92.3% and overall response rate of 100% in the per-protocol cohort.

• Three patients (17.6%) developed immune-related adverse event (grades 1–2) in neoadjuvant treatment.

Lung cancer has a high incidence and mortality rate in China and worldwide. Small cell lung cancer (SCLC) accounts for ~15% of all lung cancers¹. However, due to the rapid growth and early metastasis of SCLC, 70% of patients are diagnosed at an advanced stage. In addition, ~10–25% of patients develop brain metastases at the time of initial diagnosis, with the remaining 40–50% developing brain metastases during the course of their disease². Over the last four decades, more than 40 phase III studies and more than 60 drugs have failed in clinical trials³. Consequently, platinum (cisplatin or carboplatin) combined with etoposide-based chemotherapy remains the first-line treatment option for patients with extensive-stage SCLC⁴. However, the majority of these patients experience tumor recurrence or metastasis within a year of treatment, resulting in an extremely low mean overall survival (OS). Even in the limited-stage, patients undergoing surgery along with concurrent radiotherapy and chemotherapy also have a low OS. Therefore, developing a more effective and high-quality treatment modality is crucial.

Immune checkpoint inhibitors (ICIs) have demonstrated significant efficacy in treating non-SCLC. However, ICIs targeting extensive-stage SCLC have encountered bottlenecks. Although, in KEYNOTE-158, a phase 2 clinical trial targeting extensive-stage SCLC patients who had failed at least first-line therapy, 2 and 14 patients, respectively, attained complete remission (CR) and partial remission (PR), with an overall response rate (ORR) of 19.3%. 65% of the patients who attained remission remained in remission for more than 18 months⁵. But programmed death receptor 1 (PD-1) inhibitors combined with chemotherapy met the progression-free survival (PFS) metric of the primary endpoint and did not reach a significant OS of the endpoint^6,7. Atezolizumab, which has been approved for the first-line treatment of extensive-stage SCLC, a humanized IgG1 monoclonal antiprogrammed death receptor ligand-1 (PD-L1) antibody, inhibits PD-L1-PD-1 and PD-L1–B7-1 signaling and regains tumor-specific T-cell immunity^8–10. Horn et al. ¹⁰ were the first to demonstrate that atezolizumab combined with chemotherapy significantly improved OS and PFS in patients with advanced SCLC when compared to chemotherapy alone. According to Impower133, first-line chemotherapy combined with atezolizumab improved OS (12.3 vs. 10.3 months; P=0.007) and PFS (5.2 vs. 4.3 months; P=0.02) compared to chemotherapy alone¹¹. According to recently updated data derived from 22.9 months of follow-up, the OS of the 18 months for the combination group was 13% higher than in the chemotherapy group (33.5 vs. 20.4%). Similarly, the CASPIAN study confirmed the benefit of chemoimmunotherapy in terms of OS¹². In another study, tislelizumab combined with chemotherapy reached a median OS of 15.3 months in patients with extensive-stage SCLC¹³. The NCCN guidelines recommend surgery-based combination therapy for patients with limited-stage (T1-2N0M0) SCLC. However, neoadjuvant treatment stage II–III patients also gained surgical survival benefits¹⁴. Therefore, this study aimed to investigate the safety and efficacy of immunotherapy combined with chemotherapy for the neoadjuvant or conversion treatment of resectable SCLC.

Methods

Research design

The Department of Oncology at Beijing Chest Hospital, Capital Medical University, and the Department of Thoracic Surgery at Tangdu Hospital, Air Force Medical University conducted this multicenter, single-arm, open study to determine the safety and efficacy of ICIs in combination with chemotherapy for the neoadjuvant treatment of resectable SCLC. The clinical trial started on 15 May 2020, and ended on 15 October 2022. The protocols were approved by the Ethics Committee of Capital Medical University (Ethical Approval No. 2021-NO.4), with the clinical study registration number (ChiCTR2100042367).

Inclusion criteria

The inclusion criteria for patients were as follows: untreated limited-stage SCLC diagnosed by histopathology at an age greater than 18 years and clinical stages I–IIIB (T1-4N0-2M0) (referring to the AJCC 8th edition); SCLC without metastasis confirmed by general imageological examinations (PET/CT or enhanced CT of the chest and abdomen, MRI of the head, and radionuclide bone scan. According to the NCCN guideline, the standard for identifying abnormal lymph node stay as enlarged lymph nodes that were more than 1 cm in short axis on CT or MRI scans in patients with nonsmall cell lung cancer); there were no contraindication for surgery, which had been confirmed by cardiac ultrasound, pulmonary function, and blood gas analyses; eastern cooperative oncology group physical status score of 0–1 within 10 days prior to randomization; no contraindications to the use of ICIs confirmed by hematology laboratory tests; no other comorbid and active malignancies within the last 5 years; no Chinese herbal medicine treatment within the last 1 month; and no live vaccination within 30 days prior to the first dose of the test drug.

Exclusion criteria

The exclusion criteria for patients were as follows: patients with interstitial pneumonia; patients with systemic immune diseases receiving hormones; patients with history of HIV infection; patients with history of hepatitis B (defined as a positive hepatitis B virus surface antigen) or known active hepatitis C virus infection [defined as detection of HCV RNA (characterization)]; female patients with pregnancy detected within 24 h prior to dosing; patients with a history of organ transplantation; patients with active tuberculosis; and patients with hypersensitivity to ICIs or their excipients.

Treatment protocols

The dosing regimen for this prospective, single-arm, open study was atezolizumab or tislelizumab combined with conventional chemotherapy [carboplatin, AUC 5, D1 or nedaplatin (80 mg/m²), D1 combined with etoposide (100 mg/m²), D1–3] (Table 1). Preoperative neoadjuvant or conversion therapy was given for three cycles. Surgery was performed within 4–6 weeks of the end of the neoadjuvant treatment cycle. The anesthetic procedure was be briefly filled with: patients were given total intravenous general anesthesia and double-lumen bronchial catheter were inserted and fibreoptic bronchoscope was be used to ensure proper catheter position.

Table 1.

Baseline characteristics of included patients.

							Diameter (mm)
Patients	Sex	Age	Smoking index	Medication regimen	Neoadjuvant therapy cycles	Clinical TNM (cTNM)	After 2 cycles	After 3 cycles	RECIST 1.1 (preoperation)	Postneoadjuvant therapy TNM (ypTNM)	pCR	Postoperative treatment
P1	F	59	0	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab 1200 mg, D1	2	T2N1M0,IIB	19	19	PR	ypT0N0M0	Yes	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab 1200 mg, D1*4 Cycles +Atezolizumab 1200 mg, D1*2
P2	M	51	400	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab 1200 mg, D1	3	T2N1M0, IIB	30	21	PR	ypT0N0M0	Yes	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab 1200 mg, D1 *1 Cycles
P3	F	43	0	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab 1200 mg, D1	3	T4N2M0, IIIB	28	20	PR	ypT1aN0M0, IA	Non	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1) *2 Cycles
P4	M	51	0	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab (1200 mg, D1)	3	T2bN1M0,IIB	33	30	PR	ypT0N0M0	Yes	Etoposide (100 mg/m2,D1–D3)+ Carboplatin (AUC=5, D1)+ Atezolizumab (1200 mg, D1)*3→Atezolizumab 1200 mg, D1*5
P5	M	62	450	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	2	T1bN1M0, IIB	16	16	PR	ypT1aN0M0, IA	Non	Atezolizumab 1200 mg, D1*9
P6	M	51	450	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	3	T2aN0M0, IB	22	22	PR	ypT0N0M0	Yes	Atezolizumab 1200 mg, D1*9
P7	M	55	400	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	3	T1cN1M0, IIB	0	0	CR	ypT0N0M0	Yes	Atezolizumab 1200 mg, D1*6
P8	M	52	900	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	2	T4N1M0, IIIA	24	24	PR	ypT1cN0M0, IA	Non	Atezolizumab 1200 mg, D1*2
P9	M	45	300	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	2	T3N2M0, IIIB	21	21	PR	ypT0N0M0	Yes	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1) *1 + Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)
P10	M	59	500	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	3	T1N0M0, IA	6	3	PR	ypT0N0M0	Yes	Atezolizumab 1200 mg, D1*1
P11	M	54	0	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	4	T3N2M0, IIIB	24	18	PR	ypT0N0M0	Yes	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1) *1
P12	M	55	800	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	4	T4N1M0, IIIA	14	14	PR	ypT1cN1M0, IIB	Non	Atezolizumab (1200 mg, D1) *5
P13	F	62	0	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)	3	T3N0M0, IIB	16	12	PR	ypT1N1M0, IIB	Non
P14	M	56	0	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)		T2aN2M0, IIIA	18	13	PR
P15	M	55	800	Etoposide (100 mg/m2,D1–D3)+ Cisplatin (60 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)		T4N1M0, IIIA	52	PD	PD
P16	M	71	1000	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)		T2N2M0, IIIA	37	7	PR
P17	M	61	800	Etoposide (100 mg/m2,D1–D3)+ Nedaplatin (80 mg/m2,D1, D1)+ Atezolizumab (1200 mg, D1)		T2N2M0, IIIA	30	29	PR

CR, complete response; PD, progressive disease; pCR, pathological complete response; PR, partial response; RECIST 1.1, Response Evaluation Criteria in Solid Tumors version 1.1; SD, stable disease.

Primary endpoint

The primary endpoint of this study was pathological complete response (pCR) in the per-protocol (PP) cohort, defined as the complete absence of all tumor cells from the primary site and lymph nodes.

Secondary endpoints

1.5.2.1 Safety: adverse events (AE) (referring to common terminology criteria for adverse events version 4.0) and perioperative complications (referring to Clavien–Dindo).

1.5.2.2 Efficacy ORR calculated as the response evaluation criteria in solid tumors version 1.1(RECIST 1.1) for solid tumors: CR, complete disappearance of imaging signs; PR, target lesion shrunk by 30% or more; stable disease (SD), target lesion change (±20%); progressive disease (PD), target lesion increase more than 20% or new other lesion; major pathological response (MPR), remaining tumor cells accounted for less than 10% or primary lesion disappearance with lymph node metastasis; pathological down-staging: post-treatment ypTNM down-staging according to the TNM staging system, with no new lesions or evidence of PD; R0 resection rate; and event-free survival (EFS) or OS in the ITT or PP population.

Statistical analysis

Samples were collected based on the results of previous studies, which showed that the ORR rate of chemotherapy should be ~80–85.7%^15,16. In previous studies, the pCR after neoadjuvant chemotherapy was only 17.5%¹⁷. This study was conducted in two phases¹⁸. The first phase was divided into two groups, the tislelizumab group (TC group) and the atezolizumab group (AC group). Considering that the new treatment model was not easy to be accepted by patients and limited-stage SCLC is the low incidence rate, we used pCR as the primary endpoint and set a sample size of three in each group. After treatment, two patients in the AC group had pCR and 1 MPR, and two patients in the TC group had pCR and one non-MPR; meanwhile, the ORR was 100% in the AC group and 66.7% in the TC group. Combining the secondary endpoints of MPR and ORR, the effect of the AC group was better than that of the TC group, so the AC group was selected to enter the second-phase of the study. The sample size at the second-phase of the trail was calculated by the single-arm objective performance criteria (OPC). A hypothesis-testing evaluation was used, with the one-sided null hypothesis that pCR in the PP population was equal to or greater than OPC, the probability of type 1 error was 2.5%, and the probability of type 2 error was 30% (70% power). The pCR of 60% (according to the results of the first phase of the trial) for that OPC was chosen to produce the final minimum sample size requirement for evaluation of all OPC; that was calculated as ~10 patients in the PP patients. In practice, each pCR rate would have a confidence interval computed; the lower 95% bilateral-sided confidence limit would need to be larger than the target value (0.25, according to previous studies, the pCR of neoadjuvant chemotherapy was 17.6%¹⁷ and statistician chose 0.25 to target value) to support approval¹⁹. Several formulas have been proposed for Poisson confidence limits. The one recommended for assessing valve OPC was suggested by Clopper–Person. The work has been reported in line with the strengthening the reporting of cohort, cross-sectional and case-control studies in surgery (STROCSS) criterion²⁰, Supplemental Digital Content 1, http://links.lww.com/JS9/A761.

Continuous variables, due to noncompliance with the normal distribution, were expressed as median, whereas categorical variables were expressed as frequencies. Variables were compared using the t-test, χ²-test, analysis of variance, and Fisher’s exact test. The significance level was set to P < 0.05.

Results

Basic information on patients

From May 2020 to October 2022, a total of 25 patients were recruited (included in Phase I and Phase II), with 20 eventually being included in this study. The trial in the TC group was discontinued because the ORR criterion set for the clinical trial was not met (Appendix 1, Supplement 1, Supplemental Digital Content 2, http://links.lww.com/JS9/A762). Finally, in two phases, 17 patients, including 14 males and three females, with a mean age of 55.2 (43–71) years, were included in the AC group (Fig. 1). In addition, one, one, six, six, and three patients in this group were in stages IA, IB, IIB, IIIA, and IIIB, respectively. Ten patients had a smoking index greater than or equal to 400 (Table 1).

Figure 1.

Patients enrolled.

Clinical characteristics

Seventeen patients completed two phases of AC group in neoadjuvant or conversion therapy. Of these patients, thirteen completed the surgery, one underwent concurrent chemoradiotherapy due to PD, one dropped out of the group due to symptomatic remission, and two declined the surgery. Of the 13 patients who underwent surgery, four completed two cycles of treatment, two completed four cycles of treatment due to the epidemic, and seven completed three cycles of treatment (Table 1). In addition, the R0 resection rate in this group was 92% (12/13), with one (1 , 7.7%) case of R1 resection. Eight patients underwent minimally invasive resection (61.5%, 8/13), with five patients (5/13, 38.5%) having an open thoracotomy. In terms of surgical scope, three (23.1%), three (23.1%), four (30.8%), and three (23.1%) patients underwent pneumonectomy, sleeve resection and bronchoplasty, bilobectomy, and lobectomy, respectively. After completion of neoadjuvant chemoimmunotherapy, the mean time to complete surgery was 37.2 (28–59) days. No surgeries were delayed due to AEs, except for three patients who underwent exceeding the scheduled time for surgery due to the epidemic (COVID-19 pandemic). Moreover, patients in this group had a median surgery time of 125 (95–345) min, a median intraoperative bleeding of 200 (50–700) ml, a median hospital stay of 11 (6–16) days, and a median chest tube duration of 5 (3–12) days (Table 2). And patients had median duration of intubation and artificial ventilation of 155 (125–370) min and 145 (110–365) min.

Table 2.

Surgical information and postoperative complications.

Surgical information (N, %)
Extent of resection
R0	12 (92.3%)
R1	1 (7.7%)
Median operative time (min)	125 (95–345)
Median duration of intubation	155 (125–370)
Median artificial ventilation	145 (110–365)
Median blood loss (ml)	200 (50–700)
Median postoperative hospital stays (days)	11 (6–16)
Median chest tube duration (days)	5 (3–12)
Pathological evaluation (PP patients)
pCR	8 (61.5%)
Non-pCR	5 (38.5%)
MPR	12 (92.3%)
Non-MPR	1 (7.7%)
Extent of surgery
Lobectomy	3 (23.1%)
Bilobectomy	4 (30.7%)
Sleeve resection/bronchoplasty	3 (23.1%)
Pneumonectomy	3 (23.1%)
Surgical method
VATS/RATS	8 (61.5%)
Conversion to thoracotomy	0
Thoracotomy	5 (38.5%)
Postoperative complications
Intraoperative blood transfusion	3 (23.1%)
Death within 30 and 90 days	0
Pneumonia	2 (15.4%)
Hoarseness	1 (7.7%)
Heart failure	2 (15.4%)
Respiratory failure	0
ARDS	0

ARDS, acute respiratory distress syndrome; MPR, major pathological response; PP patients, per-protocol patients; pCR, pathological complete response; VATS/RATS, video assisted thoracic surgery/robotic-assisted thoracic surgery.

Pathological down-staging was observed in 12 patients (12/13, 92.3%) relative to preoperative clinical staging. Additionally, in the postoperative pathological assessment, 8 (8/13, 61.5%) and 12 (12/13, 92.3%) patients in the PP cohort attained pCR and MPR, respectively. pCR (8/17, 47.1%), MPR (12/17, 70.6%), and non-MPR (1/17, 5.9%) were attained in 8, 12, and 1 patient, respectively, in the intention-to-treat (ITT) cohort (Fig. 2). As assessed by RECIST 1.1, 15 (15/17, 88.2%) patients and 1 (5.9%) attained PR, and CR, respectively, in the ITT analysis (ORR of 94.1%). In addition, the mean tumor size reduction was 63.4%. The ORR in the PP cohort was 100% (13/13).

Figure 2.

Swimmer plot of EFS in the modified intention-to-treat population (n=17). Each bar represents one patient. The left column shows clinical characteristics. EFS, event-free survival.

The OS is the key indicator for assessing the efficacy of lung cancer therapies. At the time of data cutoff (15 February 2023), a mean follow-up of 21.6 (7.4–32.5) months, patients in the ITT population had not attained the median OS and EFS (Fig. 3A, B). At the same time, the median OS of the pCR patients and the median EFS of the surgical patients had not been achieved. However, the median OS of the patients of non-pCR was 18.2 months and the median EFS of the nonsurgical patients was 9.5 months (Fig. 3C, D). In addition, three patients in the surgical patients developed distant metastases, including one brain metastasis. In the nonsurgical patients, one and two patients experienced recurrence of the primary focus and distant metastases (Table 3), respectively.

Figure 3.

(A) Overall survival in the intention-to-treat population; (B) Event-free survival for intention-to-treat population; (C) Overall survival for pathological complete response and nonpathological complete response patients; (D) Event-free survival for surgery and nonsurgery patients.

Table 3.

First progression site.

Variables	Surgery people	Nonsurgery people
Relapse		1
Metastasis	3	2
Brain metastasis	1

Safety

Treatment-related AEs are shown in Table 4. Grades 1–2 AEs occurred in all patients during the neoadjuvant treatment; grade 3 or higher AEs occurred in 58.8% (10/17) of patients; and grade 4 leukopenia occurred in one patient (Table 4). In addition, immune-related adverse event (irAE, grade 2) occurred in three patients (17.6%), including one (5.9%) and two (11.8%) cases of hyperthyroidism and hypothyroidism, respectively, in the ITT population. IrAEs occurred in seven patients (7/13, 53.8%) receiving adjuvant therapy in the PP cohort, including rash, hypothyroidism, diabetes mellitus, and immune-related nephritis in three (23.1%), two (15.3%), one (7.7%), and one (7.7%) patients, respectively. Grade 3 or higher irAEs occurred in three patients (23.1%), including rash, hypothyroidism, and diabetes mellitus (Table 4). The symptoms of these patients resolved following hormone administration and symptomatic treatment. No patients died within 90 days of surgery. However, two patients developed pneumonia. These two cases of complications were not included in immune-related pneumonia because they resolved as soon as the antibiotic treatment was administered.

Table 4.

Treatment-related adverse events.

Variables	Any grade	Grades 1–2	Grade 3	Grade 4
Preoperation
Leukopenia	8	6	2
Neutropenia	7	4	2	1
Thrombocytopenia	1	1
Anorexia	5	3	2
Vomiting	3	2	1
Diarrhea	3	2	1
Fatigue	7	4	3
Alopecia	4	3	1
Hypothyroidism	2	2
hyperthyroidism	1	1
Pro-operation
Skin rashes	3	2	1
Hypothyroidism	2	1	1
Diabetes	1		1
Immune associated nephritis	1	1

Discussion

This study was conducted in two phases. In the first phase, six patients were assigned to the TC group and AC groups in a 1:1 ratio. ORR screened atezolizumab as the second-phase drug. To the best of our knowledge, this is the first study to report the application of chemotherapy in combination with atezolizumab as neoadjuvant therapy in patients with resectable SCLC. Thirteen of the seventeen enrolled patients underwent surgery with pCR of 61.5%, which is significantly higher than that of neoadjuvant chemotherapy (5.8–17.3%)^16,17. In a retrospective analysis of the role of durvalumab combined with chemotherapy as neoadjuvant therapy in SCLC by Meng et al. ²¹, none of the six patients enrolled attained pCR, which may be attributed to the use of only two cycles of neoadjuvant therapy. According to the previous studies, 3–4 cycles of therapy resulted in a higher pCR in non-SCLC than 1–2 cycles²². Clinical trials in multiple tumor types, particularly non-SCLC, have shown a greater OS benefit following neoadjuvant pCR compared to non-pCR patients²³. In addition, the prognosis of postoperative pCR patients is similar to that of patients with stage IB lung cancer in NSCLC. Therefore, pCR can be used as a valid indicator to assess the outcome of neoadjuvant therapy. However, whether a higher pCR in the immunotherapy translates into an OS benefit for patients requires further clinical RCT validation.

Surgery-based combination therapy is the preferred option for early-stage, limited-stage SCLC (T1-2N0M0). This study enrolled nine patients with locally advanced (stage III) disease for the following reasons: patients in stage III have a lower rate of local recurrence after neoadjuvant treatment with EP regimens than patients without surgery, particularly for OS benefit²⁴; according to Lewin’ski et al. ²⁵, T or N down-staging after preoperative induction chemotherapy in stage III patients, or even pCR, would result in an OS benefit. Patients included in this study had a significant reduction in the primary focus and mediastinal tumor lymph nodes following treatment. The surgery was performed according to the experimental protocol. Four (66.7%) patients attained pCR after surgery, which remained superior to chemotherapy alone²⁵; and according to CheckMate-816, the pCR of neoadjuvant immunochemotherapy for non-SCLC in III stage was 23% higher than chemotherapy alone. In addition, the neoadjuvant immunochemotherapy patients had significantly higher OS and DFS. Subgroup analysis also revealed that stage III patients had a higher OS and DFS benefit than stage I and II patients²⁶. Thus, this study investigated the efficacy of immunotherapy in stage III SCLC. However, no significant difference was found in pCR between stage III and non-III stage SCLC. However, due to the small sample size, these findings need to be validated further.

In this study, 76.9% (10/13) of patients underwent pneumonectomy, bilobectomy, sleeve resection, and bronchoplasty, a higher proportion than in previous neoadjuvant chemotherapy regimens²⁷. However, in a durvalumab-related study²¹, complex lobectomy was performed in three patients (3/6, 50%), which may be related to the predominance of SCLC as a central lung cancer and the greater tendency to invade important tissues such as the bronchi and pulmonary artery, but may not be related to ICIs.

We have kept innovating for the treatment of limited-stage SCLC. As per the cost-effectiveness analysis, tislelizumab has a significant advantage over atezolizumab. The safety and efficacy of ICIs in resectable SCLC were investigated using these two ICIs. In this study, however, only three patients were included in the ‘TC’ group, with ORR and pCR of 66.7% each (Appendix 1, Supplemental Digital Content 2, http://links.lww.com/JS9/A762). Because the ORR criterion set for the clinical trial was not met, the trial in this group was discontinued. Nevertheless, the pCR in this group remained high compared to the previous neoadjuvant chemotherapy treatment. Therefore, more evidence of efficacy and safety is required.

In the study, the ORR in the ITT cohort reached 94.1% (16/17) and met the secondary endpoint. Evans et al. designed the EP-based regimen in advanced first-line therapy and attained an ORR of 85.7%, laying the foundation for the EP regimen in the first-line treatment of SCLC¹⁶. ORRs of 60 and 79.4%, respectively, were achieved in the combined treatment groups in the IMPOWER 133 and CASPIAN study^11,28. These findings further suggest that chemotherapy in combination with immunotherapy may be an effective treatment for SCLC.

The incidence of irAE events was higher in the adjuvant treatment than in the neoadjuvant treatment, which may be due to the fact that postoperative adjuvant therapy is more likely to cause irAE. In the IMPOWER 010 study²⁹, the incidence of any irAE was as high as 52%, and the incidence of grade 3 or higher AE events was 8%. Furthermore, in the NADIM study³⁰, the incidence of irAE ≥ grade 3 was 19% in the postoperative adjuvant treatment, which was significantly higher than in the neoadjuvant treatment. In addition, according to the Japanese subgroup analysis of the IMPOWER 010 study³¹, the incidence of irAE events reached 76.4% and the incidence of grade 3 or higher AE events was 16.1%. This was similar to the results of our study.

This study also has some limitations. Firstly, the OS benefit of combination therapy was unclear due to the limited observation period. So considering such aspects, one should make conclusion about the benefits of a new treatment for SCLC regarding long-term survival with caution with limited follow-up time. Second, because of the low expression of PD-L1 in SCLC, tests for immunotherapy-related biomarkers, such as MSI and TMB, were not performed. Third, without a robust method for mediastinal lymph node evaluation, potential N3 disease, which could not be ruled out readily, could add to the heterogenicity of the study sample further.

Conclusion

In conclusion, this study suggested that the combination of chemotherapy and atezolizumab as neoadjuvant therapy may be safe and effective, with high pCR and ORR. In addition, AEs are manageable with this regimen. However, this treatment needs to be further verified in phase III clinical trials.

Ethical approval

The protocols were approved by the Ethics Committee of Capital Medical University (Ethical Approval No. 2021-NO.4).

Sources of funding

The present study was supported by grants from the National Natural Science Foundation of China (No. 81871866), Zhufeng program of Fourth Military Medical University (2019rcfcyxl), and Miaozi Talent Fund of Tangdu Hospital of Air Force Military Medical University.

Author contribution

X.Y., Z.L., and H.D.: conception and design; Z.L. and T.J.: administrative support; H.D., L.S., Y.W., C.S., Z.W., Y.N., J.Z., J.S., L.T., and J.L.: provision of study materials or patients; H.D., L.S., Y.W., and C.S.: collection and assembly of data; Y.N., J.Z., J.S., and L.T.: pathological analysis. Manuscript writing and final approval of the manuscript were done by all authors.

Conflicts of interest disclosure

The authors declare that they have no conflict of interest.

Research registration unique identifying number (UIN)

• Name of the registry: Chinese Clinical Trial Registry.

• Unique Identifying number or registration ID: Registration Number: ChiCTR2100042367.

• Hyperlink to your specific registration (must be publicly accessible and will be checked): http://www.chictr.org.cn/com/25/showproj.aspx?proj=120522.

Guarantor

Hongtao Duan, Zhe Liu, and Xiaolong Yan.

Provenance and peer review

Not commissioned, externally peer-reviewed.