Abstract
Background
We sought to determine whether surgery provides a survival benefit for patients with stage II small cell lung cancer (SCLC).
Methods
We compared patients with stage II SCLC who underwent surgical resection with those who did not using the National Cancer Database from 2004-2020. Groups were propensity-matched and overall survival difference was compared with Kaplan-Meier analysis.
Results
We identified 6797 patients with stage II SCLC who received treatment. Only 10% underwent surgery. Ninety percent of patients underwent nonsurgical treatment, with the most common treatment being chemotherapy combined with radiation (n = 4610). After propensity matching, the surgical cohort had higher overall survival: 79.1% vs 70.6% at 1 year and 33.2% vs 20.4% at 5 years (P < .001).
Conclusions
In a propensity-matched analysis of patients in the National Cancer Database with stage II SCLC, we found sustained survival benefit for patients whose treatment included surgery. This survival benefit was present for both N0 and N1 disease, but most pronounced for N0 patients.
In Short.
-
▪
There is a survival benefit for patients with stage II SCLC whose multidisciplinary care includes surgery.
-
▪
The survival benefit of surgery incorporated into care is present in both N0 and N1 disease.
Small cell lung cancer (SCLC) is a neuroendocrine tumor that comprises 15% of all types of lung cancers.1,2 It is a rapidly progressive disease with early metastases, high local recurrence rates, and poor prognosis.3 Only one-third of patients present with limited-stage disease at the time of diagnosis, but there is not a universally accepted treatment pathway for these patients.4
Surgery as a treatment for SCLC fell out of favor after a multicenter randomized controlled trial in 1969 found that radiotherapy had some survival benefit compared with resection.5 Another multicenter trial in 1994 showed no survival benefit of surgery for limited stage SCLC when performed after neoadjuvant chemotherapy and before radiation as compared with chemotherapy followed by radiation alone.6 More recently, clinicians have begun to question the applicability of these studies to the current management of SCLC. Several retrospective studies have shown that surgery, when incorporated into a multidisciplinary approach with chemotherapy, may provide some survival benefit.4,7, 8, 9, 10 The optimal treatment regimen for stage II SCLC remains unclear.
We sought to evaluate survival of patients with stage II SCLC based on treatment pathways, hypothesizing that surgery would provide survival benefit.
Patients and Methods
We queried the National Cancer Database (NCDB) for adult patients aged ≥18 years with stage II SCLC based on the American Joint Committee on Cancer 7th and 8th editions from 2004-2020. Patients with incomplete data and those who did not receive any treatment, had bilateral disease, required extrapulmonary resection, or underwent an operation without curative intent were excluded. Patients with missing data were excluded (n = 2036).
Patients were stratified into 2 groups, surgical and nonsurgical. The surgical group included all patients who underwent resection of the primary site. Patients treated with chemotherapy with or without radiation therapy, and not surgery, constituted the nonsurgical treatment group. The dataset did not indicate if chemotherapy and radiation were administered neoadjuvantly or adjuvantly. The groups were propensity-matched by clinical stage, tumor site, laterality, size, age, sex, race, and Charlson-Deyo Comorbidity Index. The primary outcome, overall survival, was compared between the 2 groups in matched analyses with Kaplan-Meier analysis. All analyses were performed using SASv9.4 (SAS Institute).
Institutional review board approval was not necessary as these data are from a public, deidentified database.
Results
Overall, 7730 patients with stage II SCLC were identified, of whom 6797 received treatment (Figure 1). Among this group, 678 (10%) underwent an operation, and 6119 (90%) underwent nonsurgical management. Within the surgery group, 148 (22%) were managed with surgery alone and 530 (78%) received chemotherapy, alone or with radiation. Of the 5796 patients who were treated without surgery, 844 (15%) had chemotherapy alone, 222 (4%) had radiation alone, and 4610 (80%) had chemotherapy and radiation. Within the nonsurgical group, 323 (5%) received other treatments including immunotherapy, hormone therapy, and those not captured by the database.
Figure 1.
Flowsheet of patients diagnosed with stage II SCLC in the National Cancer Database. (Chemo, chemotherapy; SCLC, small cell lung cancer.)
When comparing the surgical and nonsurgical cohorts based on demographic and clinical characteristics, they had similar median ages, 67 and 68 years old, respectively, and race, with 91% White and 7% Black for each (Table 1). The nonsurgical group had more women, 58% vs 50% (P < .01). In both groups the majority of patients, 83% and 84%, respectively, had Charlson-Deyo scores of 0 and 1. The nonsurgical group had slightly larger tumors, median 3.1 cm vs 2.9 cm for the surgical arm (P < .01). Similarly, the nonsurgical group had a larger proportion of higher T stage tumors, with 57% having T2-T3 tumors, compared with 50% in the surgical arm (P < .01). The nonsurgical group also had higher rates of N1 disease, 86%, compared with 78% in the surgical patients (P < .01). In looking at operations performed within the surgical group, 73% had a lobectomy, 22% had a sublobar resection, and 4.8% had a pneumonectomy.
Table 1.
Demographic and Clinical Characteristics of Patients Stratified by Treatment Approach
| Characteristic | Overall | Treatment Approach |
P Value | |
|---|---|---|---|---|
| Surgical (n = 678) | Nonsurgical (n = 5796) | |||
| Age, y | 68 (61, 74) | 67 (60, 73) | 68 (61, 74) | .02 |
| Sex | ||||
| Female | 3687 (57) | 338 (50) | 3349 (58) | <.01 |
| Race | ||||
| Black | 447 (7) | 49 (7) | 428 (7) | .86 |
| White | 5832 (91) | 612 (91) | 5220 (91) | .92 |
| Total Charlson-Deyo Score | ||||
| 0 | 3582 (55) | 350 (52) | 3232 (56) | .11 |
| 1 | 1825 (28) | 210 (31) | 1615 (28) | |
| 2 | 713 (11) | 85 (13) | 628 (11) | |
| 3+ | 354 (5) | 33 (5) | 321 (6) | |
| Primary tumor site | ||||
| Central | 419 (6) | 4 (1) | 415 (7) | <.01 |
| Upper lobe | 3293 (51) | 362 (53) | 2931 (51) | |
| Middle lobe | 299 (5) | 43 (6) | 256 (4) | |
| Lower lobe | 2033 (31) | 234 (35) | 1799 (31) | |
| Overlapping lesion | 72 (1) | 15 (2) | 57 (1) | |
| Primary site unknown | 358 (6) | 20 (3) | 338 (6) | |
| Tumor size, cm | 3.0 (2.0, 4.5) | 2.9 (1.9, 4.2) | 3.1 (2.0, 4.6) | <.01 |
| Clinical T stage | ||||
| T0-T1 | 2853 (44) | 342 (50) | 2511 (43) | <.01 |
| T2-T3 | 3621 (56) | 336 (50) | 3285 (57) | |
| Clinical N stage | ||||
| N0 | 975 (15) | 149 (22) | 826 (14) | <.01 |
| N1 | 5499 (85) | 529 (78) | 4970 (86) | |
| Laterality | ||||
| Right | 3199 (49) | 341 (50) | 2858 (49) | <.01 |
| Left | 2828 (44) | 332 (49) | 2496 (43) | |
| Midline | 25 (0.4) | 1 (0.2) | 24 (0.4) | |
| Unknown | 422 (7) | 4 (0.6) | 418 (7) | |
| Treatment | ||||
| Surgery only | 148 (2) | 148 (22) | 0 (0) | <.01 |
| Chemotherapy only | 1050 (16) | 206 (30) | 844 (15) | |
| Radiation only | 228 (4) | 6 (1) | 222 (4) | |
| Chemotherapy and radiation | 4913 (76) | 303 (45) | 4610 (80) | |
| Other therapy | 135 (2) | 15 (2) | 120 (2) | |
| Surgical procedure | ||||
| Lobectomy | 493 (72.7) | |||
| Sublobar resection | 154 (22.7) | |||
| Pneumonectomy | 31 (4.6) | |||
| Surgical outcomes | ||||
| Hospital length-of-stay | 5.0 (3.0, 7.0) | |||
| 30-d readmission | 35 (5) | |||
Values are presented as n (%) or median (interquartile range).
We created 2 propensity-matched groups, each containing 600 patients (Table 2). There were no statistically significant differences between the 2 cohorts, aside from laterality. Both had a median age of 67 years, were evenly distributed by sex, and approximately 90% of patients were White. Each group was composed of approximately 50% with T0-T1 tumors and 50% with T2-T3 tumors. Eighty percent of patients in each group had N1 disease.
Table 2.
Characteristics of Propensity Score-Matched Cohort
| Characteristic | Surgical (n = 600) | Nonsurgical (n = 600) | P Value |
|---|---|---|---|
| Demographics | |||
| Age, y | 67 (61, 73) | 67 (60, 74) | .95 |
| Female sex | 308 (51) | 315 (53) | .69 |
| Race | |||
| Black | 38 (6) | 46 (8) | .66 |
| White | 546 (91) | 536 (89) | .62 |
| Clinical Characteristics | |||
| Charlson-Deyo Score | .08 | ||
| 0 | 308 (51) | 337 (56) | |
| 1 | 184 (31) | 168 (28) | |
| 2 | 79 (13) | 57 (10) | |
| 3+ | 29 (5) | 38 (6) | |
| Clinical T stage | .82 | ||
| T0-T1 | 305 (51) | 301 (50) | |
| T2-T3 | 295 (49) | 299 (50) | |
| Clinical N stage | .94 | ||
| N0 | 118 (20) | 119 (20) | |
| N1 | 482 (80) | 481 (80) | |
| Tumor size, cm | 3.0 (2.0, 4.1) | 3.0 (1.9, 4.5) | .86 |
| Treatment | |||
| Chemotherapy only | 204 (34) | 187 (31) | .30 |
| Radiation only | 6 (1) | 8 (1) | .59 |
| Chemotherapy and radiation | 299 (50) | 306 (51) | .69 |
| Other therapy | 91 (15) | 99 (17) | .39 |
| Laterality | |||
| Right | 291 (49) | 327 (55) | .04 |
| Left | 304 (51) | 271 (45) | |
| Midline | 1 (0.2) | 2 (0.3) | |
| Unknown | 4 (0.7) | 0 (0) |
Values are presented as n (%) or median (interquartile range).
Survival in the propensity-matched surgery group for both N0 and N1 subsets was statistically significantly better than that of the respective nonsurgical groups (Figure 2). Five-year overall survival for N0 surgical patients was 44% compared with 26% for N0 nonsurgical patients (P < .001). For patients with N1 disease, 5-year overall survival was 30% of N1 surgical patients vs 19% for N1 nonsurgical patients (P < .001).
Figure 2.
Kaplan-Meier analysis of survival by N stage and treatment type in the propensity score-matched cohort. (NCDB, National Cancer Database; SCLC, small cell lung cancer.)
Comment
Patients with stage II SCLC whose management included surgery had a survival benefit when compared with those who did not undergo surgical resection. Better survival was evident at 1-, 3-, and 5-year follow-up intervals. A difference in survival was also present regardless of nodal status, although the survival improvement was more pronounced for patients with N0 disease.
Our work challenges a long-standing treatment paradigm that surgery provides no benefit for small cell lung cancer based on limited, historical data. This was first established after a multicenter randomized controlled trial in 1969 comparing surgery to radiotherapy for limited SCLC.5 Within this cohort, pneumonectomy was the sole operation offered. Median survival favored radiation therapy, 300 days compared with 199 days for surgery (P = .004). Five-year survival was 1% for surgery and 4% for radiation. There are several limitations that notably restrict the applicability of this study, including small cohort size, limited staging modalities, lack of other treatments, especially platinum-based chemotherapy, and the sole operation offered being pneumonectomy. Pneumonectomy, especially in that era, was a morbid operation associated with increased complications and mortality. A small minority of our cohort, 4.8%, underwent pneumonectomy.
Another often-cited multicenter randomized controlled trial compared survival in patients with limited-stage SCLC if surgery was incorporated into their multidisciplinary approach.6 All patients underwent chemotherapy and radiation therapy, with half of the cohort also undergoing resection. The survival curves were similar, with the nonsurgery group having a slightly better median survival at 18.6 months compared with 15.4 months in the surgery group. The authors concluded that surgery offered no benefit to the then-accepted treatment plan of chemoradiation. That study has limited applicability to dictate decisions for stage II SCLC as the authors included T3 and N2 disease. The chemotherapy was administered in the neoadjuvant setting and their regimen consisted of cyclophosphamide, doxorubicin, and vincristine, which differs from standard today.
Our study adds to mounting evidence supporting surgery in the management of stage II SCLC. Schreiber and colleagues10 performed a retrospective analysis of the Surveillance, Epidemiology, and End Results (SEER) registry and compared survival of patients who underwent surgery as part of their care vs those who did not. They found that surgery was associated with better 5-year survival, 44.8% vs 13.7% (P < .001), in patients with T1-2 disease. The 5-year survival in our propensity-matched cohort was also better in the surgery group, 33.2% vs 20.4% in the nonsurgical group (P < .001). Their work incorporates stage I disease, which likely explains the overall higher survival rates. When comparing survival among groups based on nodal status, Schreiber and associates also found a benefit to surgery for both N0-N1 disease. Median survival for the N0 surgery group was 40 months compared with 15 in the nonsurgery group (P < .001). Median survival in the N1 group was 29 months for surgery vs 14 months in the nonsurgery group (P < .001). In comparison, our N0 cohort who had surgery experienced a median survival of just over 48 months, vs 24 months in the N1 surgical cohort. In N0 patients who did not undergo surgery, median survival was just under 24 months, while median survival for N1 patients who did not undergo surgery was almost 18 months (P < .0001). These results are fairly similar (48 vs 40 months for N0 surgical patients, and 24 vs 29 months for N1 surgical patients), which underscores the benefit of surgery for both groups.
This study has several limitations. It is a retrospective analysis of a nationwide oncology outcomes database and thus is susceptible to selection bias. There are likely confounding variables we are not accounting for. The NCDB is hospital-based and not population-based, so generalizability of our findings to the global population is hindered. Data from the NCDB lacks granular patient-level variables including performance status, smoking history, chemotherapy regimens, radiation doses, specifics of surgical technique, and diagnostic modalities used, thus there is great potential for bias to exist, for which we cannot account. For patients with combination therapy, there are no data provided to determine whether chemotherapy was administered in the neoadjuvant or adjuvant setting for surgical patients. We presume it would be given as adjuvant therapy, but cannot confirm this assumption. Given the retrospective nature of this study and need for longitudinal survival of the cohort to complete our analysis, we do not have enough data for a separate analysis for the role of immunotherapy. We acknowledge that patients in the cohort receiving surgery alone for stage II SCLC with no adjuvant therapy is not standard of care. Due to limitations of the NCDB we do not have explanations for why adjuvant therapy was not delivered to this subset of patients. We excluded patients with missing data, including the treatment they underwent, so we believe this to be accurate. Given this, it is possible these patients did not recover well enough to tolerate chemotherapy or elected to forgo this component of their care. Another limitation is that in the NCDB pathology does not account for tumors with mixed biology, labeling them as 1 entity despite having 1 or more histologic characteristics. Therefore, it is possible that some treatment decisions and subsequent outcomes were based on mixed histology rather than a single histology.
In conclusion, in this retrospective study using the NCDB we performed a propensity matched analysis and found survival benefit for patients with stage II SCLC whose multidisciplinary treatment included surgery. This survival benefit was present at 1-, 3-, and 5-year intervals and was also seen in both N0 and N1 disease, although the impact of surgery appeared to be greater in N0 patients. Our work adds to a limited number of studies supporting the role of surgery in the management of stage II SCLC, which merits further investigation with prospective trials.
Acknowledgments
Funding Sources
The authors have no funding sources to disclose.
Disclosures
The authors have no conflicts of interest to disclose.
Footnotes
Presented as a Poster at the One-hundred-fourth Annual Meeting of the American Association for Thoracic Surgery, Toronto, Ontario, Canada, Apr 27-30, 2024.
References
- 1.Bernhardt E.B., Jalal S.I. Small Cell Lung Cancer. Cancer Treat Res. 2016;170:301–322. doi: 10.1007/978-3-319-40389-2_14. [DOI] [PubMed] [Google Scholar]
- 2.Rudin C.M., Brambilla E., Faivre-Finn C., Sage J. Small-cell lung cancer. Nat Rev Dis Prim. 2021;7:3. doi: 10.1038/s41572-020-00235-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Quoix E., Fraser R., Wolkove N., Finkelstein H., Kreisman H. Small cell lung cancer presenting as a solitary pulmonary nodule. Cancer. 1990;66:577–582. doi: 10.1002/1097-0142(19900801)66:3<577::aid-cncr2820660328>3.0.co;2-y. [DOI] [PubMed] [Google Scholar]
- 4.Tariq S., Kim S.Y., Monteiro de Oliveira Novaes J., Cheng H. Update 2021: management of small cell lung cancer. Lung. 2021;199:579–587. doi: 10.1007/s00408-021-00486-y. [DOI] [PubMed] [Google Scholar]
- 5.Fox W., Scadding J.G. Medical Research Council comparative trial of surgery and radiotherapy for primary treatment of small-celled or oat-celled carcinoma of bronchus. Ten-year follow-up. Lancet. 1973;2:63–65. doi: 10.1016/s0140-6736(73)93260-1. [DOI] [PubMed] [Google Scholar]
- 6.Lad T., Piantadosi S., Thomas P., Payne D., Ruckdeschel J., Giaccone G. A prospective randomized trial to determine the benefit of surgical resection of residual disease following response of small cell lung cancer to combination chemotherapy. Chest. 1994;106(6 Suppl):320S–323S. doi: 10.1378/chest.106.6_supplement.320s. [DOI] [PubMed] [Google Scholar]
- 7.Peng A., Li G., Xiong M., Xie S., Wang C. Role of surgery in patients with early stage small-cell lung cancer. Cancer Manag Res. 2019;11:7089–7101. doi: 10.2147/CMAR.S202283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Weckler B.C., Baldes N., Schirren J. Survival following multimodality treatment including surgery for stage IA-IIIB small-cell lung cancer. Thorac Cardiovasc Surg. 2019;67:291–298. doi: 10.1055/s-0038-1667145. [DOI] [PubMed] [Google Scholar]
- 9.Barnes H., See K., Barnett S., Manser R. Surgery for limited-stage small-cell lung cancer. Cochrane Database Syst Rev. 2017;4:CD011917. doi: 10.1002/14651858.CD011917.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Schreiber D., Rineer J., Weedon J., et al. Survival outcomes with the use of surgery in limited-stage small cell lung cancer: should its role be re-evaluated? Cancer. 2010;116:1350–1357. doi: 10.1002/cncr.24853. [DOI] [PubMed] [Google Scholar]