Invited Commentary. The Royalty of Evidence: The Randomized Control Trials

This article is referred to by:

Survival Outcome after Stereotactic Body Radiation Therapy and Surgery for Early Stage Non-Small Cell Lung Cancer: A Meta-Analysis

In the last issue of the journal we came across a propensity matched analysis of stage I non-small cell lung cancer (NSCLC) patients treated either by means of surgery or by stereotactic body radiotherapy (SBRT)[1]. The authors report equal local control survival, regional control survival, loco-regional control survival, distant control survival (DC), disease-free survival and progression-free survival, however SBRT patients had worse overall survival rate compared to surgical patients. This paper is interesting and also thought-provoking. Despite the in depth analysis of this paper, all the included studies were based on retrospective data. Most of the referenced papers quote survival times at 2–3 years, not the standard 5-year survival. The authors report favorable overall survival for the surgical group, irrespective of matching and support the imperative need that their conclusions should be strengthened by randomized control trials (RCTs). This report reflects the ongoing debate between SBRT and surgery.

Stage I non-small cell lung cancer (NSCLC) is potentially curable, and surgery is considered the standard of care for patients with minimal comorbidities and good performance status. Notably, the 5-year overall survival (OS) has been reported to be around 60% in some studies [2]. Nevertheless, a significant portion of these patients are poor candidates for surgery due comorbidities such as chronic obstructive pulmonary disease (COPD), cardiovascular disease or poor physical status. Stereotactic ablative radiotherapy (SABR) or stereotactic body radiotherapy (SBRT) was employed as an alternative treatment strategy for cancer patients. This method, which is a non- invasive treatment alternative to surgery, adopts modern radiotherapeutic techniques to deliver large doses of radiation and has shown superiority to conventional radiotherapy in terms of local control and toxicity [3]. Many studies have attempted a comparison between SBRT and surgery in stage I patients. To our knowledge, all of them are retrospective studies and by definition fail to provide a robust comparison. Undoubtedly, when it comes to SBRT versus surgery there is a paucity of studies to approach and examine the same kind of patients, in other words there are no RCTs. Worldwide, there have been three trials comparing surgery and SABR namely the ROSEL (NCT00687986) the STARS (NCT00840749) and the ACOSOG–RTOG (NCT01336894) [4]. All of these RCTs have closed due to inability to recruit a sufficient sample of patients. However, Chang et al, [5] have attempted a combined analysis of the ROSEL-STARS RCTs claiming that SBRT and surgery are equivalent. These authors quote that there is “valuable information in invisible and abandoned trials”. In principle we agree with this. However, the data on this paper have been incorrectly presented as level-1 evidence and the authors claim of equivalency is at least anemic. We feel that the impression given by this study, [5] that equipoise exists between SBRT and surgery is unreal.

Nonetheless, there are two RCTs trials underway which will approach and try to answer which is the best method. The first one is the veteran affairs lung cancer or stereotactic radiotherapy VALOR which originates from the USA and the second one is the SABRTooth trial which has emerged from the UK [4]. The SABRTooth trial aims to determine the feasibility and acceptability of conducting an adequately powered definitive phase III RCT comparing SABR with surgery in patients with stage I NSCLC [4]. All the retrospective studies currently available as well as the ROSEL-STARS analysis have a common denominator. They conclude that there are limitations in the number of enrolled patients and that validation with larger numbers of patients is warranted whilst they admit that only large RCTs will give the authoritative nod.

Where does the surgeon or the radiotherapist stand nowadays and what is the optimal way to proceed for the best of our patients? The results are confounding. Is there an imminent turf war between opposing camps and why the SBRT hasn't properly assessed yet? How burdensome is it to start a RCT and randomize curable patients to a treatment modality that has yet to be proved solid? All of these troublesome inquiries are valid. But since there is no compelling evidence yet, we would suggest to walk on the safe side of this treacherous path. And this side, for the time being is surgery.

The take home message here is that all scientists who are involved in this ongoing comparison, from surgeons to radiotherapists, data analyzers, statisticians, etc., need to be authentic and conscientious. Maintaining a high level of ethos is paramount to every research.

Despite the fact that this report is written by surgeons, we all acknowledge that the opportunity of a fair test should be given to less invasive treatments i.e. SBRT, but until then surgery is the only treatment modality resulting in a cure for stage I lung cancer.

Declarations of interest

The authors declare that no conflict of interest exists.