Lobectomy versus stereotactic ablative radiotherapy for medically operable patients with stage IA non-small cell lung cancer: A virtual randomized phase III trial stratified by age

Abstract

Background: Although the choice between stereotactic ablative radiotherapy (SABR) and lobectomy for early-stage non-small cell lung cancer (NSCLC) has been debated for years, the two procedures have not yet been directly compared in a randomized trial. We conducted a virtual randomized phase III trial stratified by age to compare the effectiveness of lobectomy and SABR for medically operable patients with stage IA (AJCC eighth) NSCLC using the Markov model analysis.

Methods: A Markov model was developed to simulate a cohort of patients aged 45-85 years with stage IA NSCLC who had undergone either lobectomy or SABR and were followed up for their remaining lifetime. Each virtual patient was randomly assigned to undergo lobectomy or SABR, and 10 000 patients were allocated to each group. All estimates of the variables were obtained by a systematic review of published articles.

Results: The lobectomy group showed a better life expectancy than the SABR group, in patients under 75 years of age. However, no statistically significant difference was seen in patients 75 years or older. The predicted life expectancy was 9.43 and 8.70 years in 75-year-old patients in the lobectomy and SABR groups, respectively. However, the 95%CI for the difference in life expectancy between the two groups was - 0.06-1.50 years (P = 0.0689).

Conclusions: The Markov model showed no statistically significant difference in the expected overall survival in stage IA NSCLC patients who were older than 75 years and had undergone SABR or lobectomy.

Keywords: Lobectomy; Markov model; non-small cell lung cancer; randomized trial; stereotactic radiotherapy.

Figure 1

A Scenario for the Markov state transition model of NSCLC less than 3 cm.Each rectangle represents a state of health. From the initial state, patients are randomized to undergo lobectomy or SABR. Straight arrows represent the changes that may occur during each cycle or a very short time interval. In contrast, gray rectangles mean that the patients may remain in the same Markov state for more than one cycle.NSCLC, non‐small cell lung cancer; SABR, stereotactic ablative radiotherapy; CTx, chemotherapy; RT, conventional radiotherapy; pN+, pathologically positive lymph node; NED, no evidence of disease; BSC, best supportive care.

Figure 2

Estimated overall survival stratified by age at diagnosis in patients with stage IA NSCLC after lobectomy or SABR. Patients with stage IA NSCLC were stratified by age (a) 60 years, (b) 65 years, (c) 70 years, (d) 75 years, (e) 80 years, and (f) 85 years at diagnosis and overall survival was estimated in each of the cohorts using the Markov model. () Lobectomy and () SABR.

Figure 3

Sensitivity analysis of varying probability of disease progression after primary treatment in 75‐year‐old patients. One‐way sensitivity analysis of varying probability of disease progression after (a) lobectomy and (b) SABR. SABR could be a preferred strategy if the probability of these two variables were changed beyond the threshold. () Lobectomy and () SABR. (c) Two‐way sensitivity analysis of the probability of disease progression after lobectomy and SABR: the dark gray region denotes lobectomy is preferred, while the light gray region shows SABR is preferred. SABR, stereotactic ablative radiotherapy; Dz_prog, disease progression.

Figure 4

Validation of the Markov model. The predicted (a) fiveyear survival curve after lobectomy and (b) threeyear survival curve after SABR based on the Markov model are shown. The gray circles represent the survival outcomes from real studies, wherein the median age of each cohort was set to a representative value at diagnosis. The black square is the average of the values from these real studies. The mean overall survival following lobectomy and SABR in the real studies were approximately 1% and 2% lower, respectively than those obtained from the Markov model. SABR, stereotactic ablative radiotherapy. () Markov model, () clinical trials, and () mean value of clinical trials.