Effectiveness of Thermal Ablation and Stereotactic Radiotherapy Based on Stage I Lung Cancer Histology
Is there a difference in effectiveness of thermal ablation (TA) and stereotactic body radiotherapy (SBRT) as initial treatments for stage I lung cancer depending on histologic subtypes?
Take away point
Overall survival following TA and SBRT for stage I lung cancer is comparable for most histologic subtypes, except for non-small cell neuroendocrine tumors (NETs), where overall survival is higher after TA.
Uhlig J, Mehta S, Dendy Case M, Dhanasopon A, Blasberg J Homer R, Solomon S, Kim H. Effectiveness of Thermal Ablation and Stereotactic Body Radiotherapy Based on Stage I Lung Cancer Histology. J Vasc Interv Radiol. 2021; 32:1022-1028. doi.org/10.1016/j.jvir.2021.02.025
Click here for abstract
Retrospective, database analysis using 2004-2016 data from the National Cancer Database.
No reported funding.
National database study via the National Cancer Database; USA.
The effectiveness of thermal ablation (TA) and stereotactic body radiotherapy (SBRT) according to histologic lung cancer subtypes.
Histologic subtypes of lung cancer are known to affect initial treatment and prognosis. In patients who are not surgical candidates or who have an unfavorable comorbidity profile for chemotherapy, thermal ablation (TA) or stereotactic body radiotherapy (SBRT) are options, but comparative effectiveness data is lacking. The authors perform a retrospective database analysis comparing effectiveness of TA and SBRT based on histologic subtype for initial treatment of stage I lung cancer.
The National Cancer Database was queried for patients having undergone initial treatment for stage 1 lung cancer with TA or SBRT from 2004-2016. Patients were excluded if they were <18 years of age, received surgery or chemotherapy, had unknown survival or follow up, or had a lung cancer histologic subtype that did not fall under the 2015 World Health Organization (WHO) classification system. The Charleston Comorbidity Index (CMI) was used to stratify comorbidities.
Primary outcome was overall survival defined as time from treatment until patient death or censoring. Initial data were evaluated and nearest-neighbor propensity score matching was completed with a 1:5 ratio of patients receiving TA versus SBRT. Propensity score matching was verified with a Kolmogorov-Smirnov test.
A total of 28,425 patients met criteria with 947 having undergone TA and 27, 478 having undergone SBRT. Propensity score matching was completed using the aforementioned variables identified in the multivariable logistic regression. Matched cohorts included 817 patients receiving TA and 4,085 patients receiving SBRT. Subgroup analyses were stratified according to histologic subtype as defined by the WHO 2015 classification system and included adenocarcinoma, non-NET large cell carcinoma, small cell carcinoma, non-small cell NET, and squamous cell carcinoma. Overall survival curves were generated via Cox regression and hazards ratios were calculated.
From initial analyses using the full nonmatched cohorts, TA was independently associated with Caucasian race, a greater number of comorbidities, earlier year of cancer diagnosis, smaller tumor diameter, tumor location in the lower lobe, facility location in the Middle Atlantic or New England regions, and tumor subtype of non-small cell NETs.
There was no significant difference in overall survival between TA and SBRT overall (HR=1.07, p=.13). In the subgroup analysis, OS was significantly longer for TA versus SBRT in patients with non-small cell NETs (HR=0.48, p=.04). There was no statistically significant difference in OS in the other histologic type subgroups [adenocarcinoma (HR=1.08, p=.20), non-NET large cell carcinoma (HR=2.12, p=.25), small cell carcinoma (HR=.93, p=.82), and squamous cell carcinoma (HR=1.12, p=.12)].
The authors discuss why TA may be more effective in non-small cell NETs. NETs are often well-defined, smaller diameter, and thus a better target for TA than other subtypes. Additionally, NETs are known to be less aggressive than other subtypes and are comparatively radioresistant.
While SBRT is considered the standard of care in patients with NSCLC and less invasive than TA, it often incurs longer lasting complications such as fatigue, pneumonitis, and chest wall pain. In comparison, TA has known risks of pneumothorax and pleural effusion, but can be completed on a single session, which is more convenient for the patient and more cost effective.
The authors evaluate the effectiveness of TA versus SBRT as initial treatment for patients with stage I lung cancer among histologic subtypes using data from the National Cancer Database. Several limitations exist, most inherent to the database design. Data on combination or repeat/salvage treatment, details of procedural complications, and cancer-specific recurrence were not available. This precluded evaluation of complication rate, as well as limits the overall specificity of the conclusions. Additionally, lack of technical details (such as probe count and type of ablation) limit further comparison between TA and SBRT. Lastly, there are intrinsic limitations to a retrospective study design with propensity score matching; a prospective, randomized trial would certainly be a more robust evaluation.
The main advantage of this study is the large-scale nature allowing for ample power and improved reliability of the primary endpoint that showed increased OS in patients with non-small cell NETs receiving TA versus SBRT. Interestingly, the non-small cell NET subtype was independently associated with TA, suggesting that providers may already more strongly favor TA as initial treatment in this subgroup.
The authors also offer commentary on their results regarding the demographic factors associated with TA. Despite broad results showing comparable effectiveness of TA and SBRT, TA is disproportionally utilized in the northeastern US and in Caucasian patients. The authors hypothesize this may be related to differences in socioeconomic status, insurance status, or comorbidity profile, highlighting known healthcare disparities in the US. A lack of provider awareness of TA may also contribute to this distribution.
While intrinsically limited, this large-scale study still offers a valuable insight into the effectiveness of TA and SBRT in treating the histologic subtypes of stage I lung cancer and suggests considering TA over SBRT as initial treatment in nonsurgical patients with stage I non-small cell NETs. This study also raises awareness of TA as an effective alternative treatment for stage I lung cancer in the appropriate clinical setting.
Catherine (Rin) Panick, MD
Resident Physician, Integrated Interventional Radiology
Dotter Interventional Institute
Oregon Health & Science University