Does Stereotactic Body Radiation Therapy Revolutionize the Treatment of Metastatic Non-Small Cell Lung Cancer?

H. Jack West, MD


January 14, 2015

Phase II Trial of Stereotactic Body Radiation Therapy Combined With Erlotinib for Patients With Limited But Progressive Metastatic Non-Small Cell Lung Cancer

Iyengar P, Kavanagh BD, Wardak Z, et al
J Clin Oncol. 2014;32:3824-3830

On the surface, the results sound so remarkable that the work could represent a true breakthrough: Chemotherapy pretreated, molecularly unselected patients with advanced non-small cell lung cancer (NSCLC) and up to five metastatic foci of disease had a median progression-free survival (PFS) of 14.7 months and median overall survival (OS) of 20.4 months when given stereotactic body radiation therapy (SBRT) to sites of metastatic disease along with concurrent daily erlotinib (Tarceva®).[1] With the median OS for second- or third-line erlotinib established as 6.7 months based on a large randomized trial,[2] it is clear that these results are far superior to any results we could expect to see in patients with previously treated advanced NSCLC. With SBRT increasingly available and feasible for patients whether they have isolated disease or not, is it time to fundamentally reconsider the role of local therapy for advanced NSCLC?

Despite these very encouraging numbers and their prominent publication by Iyengar and colleagues in the Journal of Clinical Oncology,[1] I believe it is appropriate to question long-held assumptions about cancer treatment but not yet consider the question answered. Specifically, these results from 24 patients at two centers should not change practice.

We have long known that a minority of patients with one or a few metastatic lesions may enjoy prolonged survival suggestive of a potential cure if both primary tumor and solitary metastatic focus are resected, particularly if the metastatic lesion is located in the brain[3] or adrenal gland.[4] In recent years, the potential utility of SBRT has transitioned from primarily being employed for medically unresectable early-stage NSCLC to a potential option for local treatment of one or several metastatic lesions with typically very minimal morbidity.[5]

That SBRT is generally safe and feasible is already quite clear. The question is no longer whether it can be done but whether it should be done. The literature in the field of oligometastatic disease is replete with single-center retrospective case series that clearly demonstrate that highly selected patients—specifically those with a low tumor burden, one or few sites of metastatic spread, and with a long prior period of disease control—can enjoy an encouragingly long PFS and often OS in absolute terms relative to a much broader historical control group of patients who are not limited by tumor burden, extent of metastatic spread, or degree or duration of prior disease control.

Patients with metastatic disease may demonstrate spread at existing or new sites of disease. A central tenet of management of metastatic NSCLC, and most other forms of metastatic cancer, is that there is not clear value in treating metastatic foci of disease with local therapy to improve survival. This is based on the presumption that survival is generally limited by the ongoing susceptibility to distant spread; if so, PFS is likely to be a meaningless goal for local therapy. Surgery can render an advanced NSCLC patient with eight metastatic lesions disease-free by scans a month later, but this shouldn't suggest that this intervention has helped the patient if, as is overwhelmingly likely, that patient has multiple new lesions apparent by the time they recover from the surgery. SBRT has the advantage of treating multiple foci of disease with minimal morbidity, but it has yet to be demonstrated that it is more effective than using Adobe Photoshop to erase the lesions on scans and merely pretend that they no longer exist. PFS is a meaningless endpoint in this setting.

Over and over, as the definition of oligometastatic disease drifts further from the prior concept of a solitary precocious metastatic lesion, now sometimes up into the range of five or six or more foci of disease, the results illustrate that as you undermine the selection process by including patients with what should be considered polymetastatic or simply frankly metastatic disease, the favorable results are overwhelmingly seen in the patients for whom the underlying biology of the disease would be expected to lead to the most favorable outcomes, regardless of the treatment administered.

Though some have estimated that liberal parameters for oligometastatic disease could be applied to at least half of the patients with metastatic NSCLC,[6] it is worth underscoring that the report by Iyengar and colleagues enrolled just 24 patients from two very large academic cancer centers over a 6-year period, several of whom were diagnosed more than 2 years before enrollment. Despite the fact that the trial allowed for patients to have up to five metastatic foci, two thirds of the patients had one or two lesions, and all but three had three or fewer metastatic lesions. This was clearly a very selected subpopulation for a treatment that is argued as being potentially broadly applicable as local therapy in metastatic NSCLC.

This is not to say that a subset of patients do not benefit from considering local therapy for very limited disease. However, single-arm trials are so clearly subject to profound selection bias that it is inappropriate to compare their results to those of a general population that has not been subject to this bias. Instead, the growing array of provocative results from small studies of SBRT highlight the great need for randomized trials of a well-defined population, as are already being conducted.[7]

We anxiously await the results of randomized trials, which hold the promise of demonstrating that metastatic disease is truly more of a continuous than a discrete variable. If so, it would refine our understanding of the potential value of local therapy for distant disease. Until that time, SBRT and other local therapies may well have a role for a limited subset of patients with symptoms or survival driven more by a local than systemic process, but we should remain very judicious about recommending it simply because it is available and the studies of it have preferentially enrolled the patients most likely to do especially well regardless of the treatment they receive.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.