Key points

Introduction

Early stage non–small cell lung cancer (NSCLC) is a growing clinical entity with evolving standards of care. As a whole, NSCLC is the second most common malignancy in men and women, with 224,390 new cases and 158,080 deaths predicted in 2016. It remains the leading cause of death from cancer, with a 5-year overall survival of approximately 21%. Deaths attributable from lung cancer are driven primarily from advanced disease, with early stage disease considered potentially curable. Although most patients have traditionally presented with late stage disease, the population of patients with early stage lung cancer is now expected to increase significantly due to improvements in diagnostic modalities and the widespread adoption of lung cancer screening programs.

In 2014, lung cancer screening with low-dose chest computed tomography (CT) was recommended for high-risk individuals by the US Preventive Services Task Force based on several large, population-based studies. The largest US-based trial was the National Lung Screening Trial, a national study accruing more than 50,000 patients demonstrating reduced lung cancer mortality with low-dose CT screening. In this study of high-risk individuals with 30 or more pack-year smoking history between the ages of 55 to 74 years of age, the relative risk of mortality from lung cancer decreased by one-fifth. Based on these results, several societies and national organizations have endorsed guidelines for lung cancer screening. The adoption of lung cancer screening guidelines is expected to significantly increase the detection of early stage disease. As the incidence of early stage lung cancer increases, it will be critical to better define appropriate treatment options for patients in this population.

Introduction

Early stage non–small cell lung cancer (NSCLC) is a growing clinical entity with evolving standards of care. As a whole, NSCLC is the second most common malignancy in men and women, with 224,390 new cases and 158,080 deaths predicted in 2016. It remains the leading cause of death from cancer, with a 5-year overall survival of approximately 21%. Deaths attributable from lung cancer are driven primarily from advanced disease, with early stage disease considered potentially curable. Although most patients have traditionally presented with late stage disease, the population of patients with early stage lung cancer is now expected to increase significantly due to improvements in diagnostic modalities and the widespread adoption of lung cancer screening programs.

In 2014, lung cancer screening with low-dose chest computed tomography (CT) was recommended for high-risk individuals by the US Preventive Services Task Force based on several large, population-based studies. The largest US-based trial was the National Lung Screening Trial, a national study accruing more than 50,000 patients demonstrating reduced lung cancer mortality with low-dose CT screening. In this study of high-risk individuals with 30 or more pack-year smoking history between the ages of 55 to 74 years of age, the relative risk of mortality from lung cancer decreased by one-fifth. Based on these results, several societies and national organizations have endorsed guidelines for lung cancer screening. The adoption of lung cancer screening guidelines is expected to significantly increase the detection of early stage disease. As the incidence of early stage lung cancer increases, it will be critical to better define appropriate treatment options for patients in this population.

Treatment options for early stage non–small cell lung cancer

For decades, the historical standard for early stage NSCLC was surgical extirpation. Similar to early treatment paradigms for breast cancer, the best chance of cure was thought to be achieved through radical surgery. As a result, the standard of care endorsed by nearly all oncologic and surgical societies was centered around surgery as the cornerstone of treatment. To date, surgery remains the favored approach, as current National Comprehensive Cancer Network guidelines recommend complete surgical resection as the preferred therapy for localized disease in patients who can tolerate the procedure.

However, the role of surgery as the de facto definitive approach has increasingly come under inquiry as the evidence has not yet been established through randomized trials. Three randomized phase 3 trials have been attempted to directly compare surgery versus radiation, but all have prematurely closed due to poor accrual. Consequently, there continues to be considerable debate as to what constitutes standard treatment options. The evidence for the use of surgery primarily relies on principles of oncologic management and early outcome data from selected groups of patients in surgical series. Furthermore, for many years, the only viable treatment option to address gross disease was limited to surgery. Historical delivery of radiation often yielded poor outcomes using conventional fractionated radiotherapy. The use of large treatment fields due to targeting uncertainty limited the ability to deliver truly therapeutic doses, resulting in toxic treatment with poor local control rates. Similar to data demonstrating potential harm in postoperative lung patients, the delivery of antiquated thoracic radiation has traditionally resulted in an unfavorable therapeutic ratio. Consequently, radiation was primarily reserved in the cases of palliation. However, as technological advances rapidly came into adoption and radiation treatment techniques became more sophisticated, effective local control of disease became increasingly achievable. As the population of early stage NSCLC continues to grow, elucidating the role of radiation therapy in the definite management of early stage disease will become a more salient issue.

Development of stereotactic ablative radiation therapy

The development of key techniques in stereotactic delivery has transformed the treatment of early stage lung cancer over the past 2 decades. Beginning in the late 1980s, several investigators across the globe began experimenting with the use of larger and more targeted doses of radiation to extracranial targets based on techniques implemented from intracranial radiosurgery. Initially deriving its nomenclature from these intracranial techniques, the original term stereotactic body radiation therapy (SBRT) refers to the use of stereotaxy to improve treatment targeting by linking orthogonal Cartesian coordinates in an external reference frame to internal cross-sectional anatomy. By doing so, the spatial accuracy and reliability of treatment delivery is substantially increased. One of the first experiences using stereotactic treatment was published from the Karolinska Hospital, where the Gamma knife radiosurgery platform was first successfully used. Data from Blomgren and colleagues first demonstrated the safety and feasibility of extracranial high-dose stereotactic treatment. In this initial experience, 31 patients with extracranial solitary tumors of the liver, retroperitoneal space, and lung were treated with stereotactic radiotherapy with doses ranging from 7.7 to 30 Gy per fraction with highly inhomogeneous, conformal treatments. Local control rates were higher than expected, with no evidence of local progressive disease in 80% of treated lesions during a relatively short follow-up period of 1.5 to 38 months. Impressively, half of the treated tumors were noted to regress or disappear. In parallel fashion, investigators at the National Defense Medical College in Japan began to investigate similar techniques. In order to account for extracranial motion, Uematsu and colleagues developed a prototype fusion of CT and linear accelerator so that positioning and targeting could be confirmed before each treatment session. Using this setup, investigators delivered high-dose stereotactic treatment to 45 patients with primary or oligometastatic primary tumors of the lung. The ranges of doses in this study were much higher than those delivered in the Karolinska experience, with Uematsu and colleagues using doses ranging from 30 to 75 Gy in 5 to 15 fractions. Given the higher doses delivered, researchers were able to achieve very high local control rates, with local progression only occurring in 2 of 66 treated lesions.

The excellent clinical outcomes from these series suggested the feasibility and rational use of stereotactic radiation as a definitive treatment option for well-circumscribed extracranial targets. In addition, given that local control rates were higher than expected based on historical radiobiological models, there was suggestion that there was unique biology occurring with significantly larger fraction sizes.

Biologic rationale for stereotactic ablative radiation therapy

Traditionally, the effects of fractionated radiation therapy have been attempted to be modeled for decades using a mathematical formula described as the linear-quadratic formula. By historical convention, the typical doses to produce this model often range from 1.8 to 8.0 Gy, with limited data incorporating the higher doses used in modern stereotactic treatments. This model was empirically derived based on in vitro data beginning in the early 1950s. Since then, experimental evidence has suggested that the predictive value of this model decreases as dose increases, with poor estimation of effects from doses delivered in the stereotactic and radiosurgical ranges. With stereotactic delivery, the goals of treatment biologically differ from that of conventional radiation therapy. Conventional fractionated radiation therapy is often delivered daily over a protracted course, relying on the cumulative differential damage of tumor tissue unable to repair damage at the same rate of normal tissue. In this manner, large amounts of normal tissue can be included in the treated field owing to potential repair mechanisms. In contrast, the goal of Stereotactic Ablative Radiation Therapy (SABR) is to ablate the targeted tissue while minimizing the normal tissue included, effectively ignoring the differential effect of radiation on normal and tumor tissue. These biologic differences are manifested in the size and dose of radiation delivered. Although conventional radiation is intended to cover large areas of normal tissue to account for subclinical disease, SABR is focused on solely gross tumor with minimal margins minimizing exposure to normal tissues.

Clinical series

Because of the historical preference of surgical therapy as the definitive therapy for early stage lung cancer, most data regarding the use of radiation have been in the inoperable or not medically fit population. As a result, survival in these series may be limited due to a negative selection bias of patients with more comorbid conditions. However, as adoption of this technology has increased, its use has become more ubiquitous.

The first North American experience using SABR was established by Timmerman and colleagues at Indiana University in a series of phase 1 and 2 trials. In the initial phase 1 dose escalation toxicity, 47 patients with inoperable early stage NSCLC were immobilized in a stereotactic body frame and treated in escalating doses of radiotherapy beginning at 24 Gy total (3 × 8 Gy fractions) and cohorts dose escalated by 6.0 Gy total. Local control in this series was excellent with minimal toxicity reported. In the phase 2 trial using the established dosing from the phase 1 study, local control remained excellent with 95% control at 17.5 months. However, it was noted that 6 patients had treatment-related deaths, possibly due to central airway necrosis from treatment.

Based on this initial experience, the multi-institutional RTOG 0236 accrued 59 patients with inoperable early stage, peripheral tumors. In this study, only 1 patient had a primary tumor failure with an estimated 3-year primary tumor control rate of 97.6%. The median overall survival was 48.1 months, which compared very favorably to historical surgery and radiation controls. No grade 5 treatment-related adverse events were reported, and no central lesions were included on this trial.

Since that time, the RTOG (now NRG Oncology) has investigated the use of SABR in several different settings for early stage patients. In RTOG 0915, investigators compared 2 treatment schedules in medically inoperable patients to determine the toxicity rate of different fractionation schemes. Patients were randomized to receive either 34 Gy in 1 fraction (arm 1) or 48 Gy in 4 consecutive daily fractions (arm 2). The 1-year local control rate was 97.0% (95% confidence interval [CI] 84.2%–99.9%) for arm 1 and 92.7% (95% CI 80.1%–98.5%) for arm 2, with no differences in toxicity in either group. In RTOG 0618, investigators studied the use of SABR in operable patients as determined by an independent thoracic surgeon using specific criteria. Although only published in abstract form, this small study of 26 patients demonstrated a 2-year local and lobar failure rate of 19%, which was higher than expected.