A Decision-Making Model for Selecting Surgery or Radiotherapy for Early-Stage Lung Cancer

Minimally invasive surgery (MIS) and stereotactic body radiotherapy (SBRT) are both used to treat stage 1 non-small cell lung cancer (NSCLC). While MIS remains the standard of care for patients with operable disease, selecting which patients should be referred for SBRT has been challenging because there are many clinical factors affecting the decision-making process.

A new study by a team of 12 thoracic surgeons at Memorial Sloan Kettering Cancer Center (MSK), published recently in the Annals of Surgery, ⁽¹⁾ is the first to identify a comprehensive list of clinical factors significantly associated with referring patients with stage 1 NSCLC to SBRT. The factors included age, reduced performance status, previous pulmonary resection, MSK-Frailty score, forced expiratory volume in one second (FEV1), and diffusion capacity of the lung for carbon monoxide (DLCO).

The investigators used these factors to create a prediction model for determining which patients with stage 1 NSCLC were most likely to benefit from referral to SBRT.  ⁽¹⁾

“The clinical factors we identified are already collected by most thoracic surgeons and pulmonologists worldwide,” said MSK thoracic surgeon Gaetano Rocco, MD, senior author of the study. “However, the model we created translates our clinical experience at MSK into a valuable tool for guiding referral decisions by categorizing patients into three distinct risk groups.”

Treatment Options for Early-Stage Lung Cancer

Surgery remains the standard of care for operable patients, while SBRT is more frequently recommended for patients deemed high-risk for surgery or who decline surgery.

Previous randomized clinical trials directly comparing surgical resection to SBRT were closed early due to poor accrual. ⁽²⁾ A pooled analysis among 58 patients from these studies showed a comparable recurrence-free survival but an improved overall survival (OS) for SBRT in patients with operable stage 1 NSCLC. However, the majority of patients in the surgical cohort had open surgery, not MIS, and 7% of patients had disease progression at the time of surgery.  ^(2) (3) 

In prior studies based on real-world clinical settings, SBRT demonstrated that it provides excellent short-term morbidity, mortality, and tumor control. While the results showed recurrence-free survival and OS were worse after SBRT compared to surgery overall, treatment modality was no longer associated with survival after adjusting for age and performance status. ^{(4) (5) (6)} 

Finally, while most patients tolerate MIS and SBRT well, postoperative complications or radiation-induced pneumonitis and esophagitis, respectively, can negatively affect the quality of life and should be considered when determining the optimal treatment modality.  ^{(7) (8) (9) (10)}

Study Design

Dr. Rocco and colleagues analyzed patient records for all consecutive patients who underwent MIS or SBRT for confirmed stage 1 NSCLC at MSK from January 2020 to July 2023. They excluded those who received neoadjuvant chemotherapy, open surgery, had cancer diagnosed preoperatively, or were surgical candidates who opted for SBRT. In total, their analysis included 1,291 patients. ⁽¹⁾

The investigators included patient demographics and characteristics, disease stage, pathology results, post-procedural outcomes, survival, and recurrence data. They also looked at the MSK-Frailty Index, a composite measure of the following clinical risk factors: chronic obstructive pulmonary disease (COPD), acute myocardial infarction, congestive heart failure, hypertension, peripheral artery disease, coronary artery disease, stroke, transient ischemic attack, diabetes, cognitive impairment, and reduced activities of daily living.  ⁽¹¹⁾

Post-procedural outcome data included 90-day mortality and surgical complication rates for surgical patients. For patients who received SBRT, the analysis included 90-day mortality as well as new-onset or treatment-induced toxicity, including cough, dyspnea, chest wall pain, pneumonitis, pulmonary fibrosis, bronchial stricture, esophagitis, dysphagia, myocardial infarction, pericarditis, and radiation dermatitis.  ⁽⁸⁾

The study’s primary objective was to identify clinical factors that were significantly associated with SBRT referral. Dr. Rocco and colleagues then used these factors to develop a predictive model. They evaluated the model’s performance and calculated two cutoff values for stratifying patients into three categories based on perioperative risk and postoperative complications. The first cutoff identified the low-risk patients most likely to be referred to MIS, and the second cutoff identified high-risk patients most likely to be referred to SBRT. Patients falling between these two cutoff values were categorized as intermediate-risk.  ⁽¹⁾

The study’s secondary objective was to validate the model’s performance by assessing post-procedural outcomes, recurrence, and OS and compare results for patients in the MIS and the SBRT groups.  ⁽¹⁾

Study Results

Among 1,291 patients treated for stage 1 NSCLC, 1,116 (86%) underwent MIS, and 175 (14%) received SBRT. Patients treated with SBRT were older (77 versus 70 years) and had worse performance status, lower pulmonary function, and higher MSK-Frailty scores than those in the MIS group.  ⁽¹⁾

In the MIS group, 582 patients (52%) underwent lobectomy or bilobectomy, 260 (23%) underwent segmentectomy, and 274 (25%) underwent wedge resection. Also, 52 patients (5%) required conversion to open surgery, and 13 (1%) had an R1 resection upon final histopathology. In the SBRT group, the median total dose was 50 Gray (Gy), most often given as 10 Gy in five fractions. All SBRT patients completed their entire treatment courses.  ⁽¹⁾

On multivariate analysis, factors associated with SBRT selection included in the decision-making model were as follows: age, performance status of 2 or 3, previous pulmonary resection, MSK-Frailty score, FEV1, and DLCO.  ⁽¹⁾

The prediction model created with these variables demonstrated an area-under-the-curve (AUC) of 0.902. Dr. Rocco and colleagues used the model to stratify 1,197 patients with calculable probability scores into three risk categories as follows:  ⁽¹⁾

• 998 patients in the low-risk category (MIS = 970, SBRT = 28),
• 149 patients in the intermediate-risk group (MIS = 96, SBRT = 53), and
• 50 patients in the high-risk group (MIS = 10, SBRT = 40).

Within the intermediate-risk group of 149 patients eligible for either treatment modality, factors significantly associated with SBRT selection on multivariate analysis included hypertension (odds ratio (OR) = 0.39, p = 0.042) and COPD (OR = 3.09, p = 0.009). However, only age was identified as a predictor of OS, with a hazard ratio of 1.2 (p = 0.021).  ⁽¹⁾

Outcomes after MIS and SBRT reflected the distinct characteristics of patients in each group. The rate of postoperative complication after MIS was 28%, and the rate of radiation-induced toxicity was 29% (p = 0.71). Rates of grade 3 or higher complications were similar — 4% for MIS and 3% for SBRT. ⁽¹⁾

90-day mortality rates were also similar at 0.4% for MIS and 0.6% for SBRT (p = 0.58). Notably, the majority of deaths in both the SBRT (23 of 31 or 74%) and MIS groups (30 of 42 or 71%) died from causes unrelated to lung cancer. Within the intermediate-risk group, three-year OS was comparable at 83% for SBRT and 91% for MIS (p = 0.60). ⁽¹⁾

“While our study did not have sufficient power to elucidate specific clinical variables associated with SBRT selection for intermediate-risk patients, we are currently prospectively collecting data for this cohort to address this gap,” said Dr. Rocco.

Main Takeaway

“The choice of MIS or SBRT was not associated with OS in the overall cohort or the intermediate-risk group, suggesting that comparable outcomes are achievable for either treatment modality with careful patient selection,” Dr. Rocco said. “Determining optimal treatment plans for intermediate-risk patients requires collaboration between multidisciplinary experts at high-volume centers, like MSK.”

Learn more about lung cancer clinical trials at MSK.

This study was partly supported by the National Institutes of Health/National Cancer Institute Cancer Center Support Grant (P30 CA008748), and MSK’s Fiona and Stanley Druckenmiller Research Center for Lung Cancer Research. Access disclosures for Dr. Rocco.