Radiofrequency ablation in high‐risk stage I non–small cell lung cancer

For many years, radiofrequency ablation (RFA) has been a part of the conversation on treatment options for patients with early-stage non–small cell lung cancer (NSCLC) who cannot tolerate lobectomy despite a paucity of prospective data on its efficacy. Proponents of RFA praise its ease of use, single-setting treatment, tolerability, and cost-effectiveness, but to date, the majority of the data for its use in the lungs has been retrospective and from single institutions or industrysponsored work. Most series also have combined early-stage NSCLC with lung metastases and have focused almost entirely on extremely short-term outcomes. In this issue of Cancer, Dupuy et al report the first multi-institutional, prospective evaluation of the technology specifically for medically inoperable early-stage NSCLC. The investigators should be congratulated on a well-executed trial. Fifty-one cT1N0 patients from 16 institutions were evaluated, and all had biopsy-proven disease, were deemed medically inoperable by a thoracic surgeon, and had clinical follow-up through 2 years; none of this is easy to accomplish in a high-risk population. Dupuy et al demonstrated excellent safety, tolerability, and preservation of pulmonary function after RFA of the lungs in a frail group of patients. Unfortunately, in this prospective evaluation, the local recurrence rate after RFA was 40% at 2 years. This is consistent with other series using RFA for early-stage NSCLC but does not compare favorably with the local control reported with stereotactic body radiotherapy (SBRT): (9.4% in-lobe recurrence at 3 years in the Radiation Therapy Oncology Group 0236 trial, which included tumors up to 5 cm) or sublobar resection (7.7% local recurrence at 3 years in the American College of Surgeons Oncology Group Z4032 trial [Alliance]). Local control was marginally better for tumors< 2 cm, but it was still not close to the rates reported for SBRT or sublobar resection. An interesting note from this trial is that local recurrence did not appear to have a detrimental impact on overall survival at 2 years. This has been reported in other series examining RFA in high-risk, early-stage NSCLC patients, and competing comorbidity clearly plays a key role in this finding. Despite the far better local control reported in the Radiation Therapy Oncology Group 0236 trial, overall survival at 3 years was only 56%, which is not very different from what is reported here 2 years after RFA. Dupuy et al also argue that the ability to re-treat a local recurrence with additional RFA, conventional external-beam radiation, or SBRT is an important reason that mortality is not negatively affected, but one should note that the follow-up in this study was quite short and question whether local failure will affect overall survival when patients are followed through 5 years. One also has to recognize that local tumor recurrence negatively affects a patient’s quality of life and the cost-effectiveness of the technology. A recent cost analysis using a Markov model of highrisk, medically inoperable NSCLC compared RFA, SBRT, and conventional external-beam radiation and found SBRT to be the most cost-effective treatment option, with an incremental cost-effective ratio of $14,000 per quality-adjusted life year over RFA and with an incremental cost-effective ratio of $6000 per quality-adjusted life year over conventional external-beam radiation. The increased incremental cost-effective ratio over RFA was primarily driven by the improved local control with SBRT, but this analysis did not take into account the ability to re-treat locally recurrent disease. The relatively high local recurrence rate after RFA does not mean that it does not have a place in the treatment of high-risk patients with early-stage NSCLC. Cancer treatment is always a balance between efficacy and tolerability, and in medically high-risk patients with NSCLC, the precise spot that appropriately balances risk and benefit is incredibly narrow, and small shifts in either direction can dramatically affect overall survival. The high-risk population is diverse in terms of both level frailty and treatment expectations. There is currently not one treatment that represents the best choice for all patients, but rather there are several options with varying levels of efficacy and tolerability and clinical indications.