A Phase I/II Trial of Stereotactic Body Radiation Therapy

Stage I Non-small Cell Lung Cancer Lung cancer is the most frequent cause of cancer death in both men and women in North America, accounting for approximately 13% of all cancers diagnosed and 28% of all cancer deaths. There will be an estimated 173,770 new lung cancer cases in the United States in the year 2004 with an estimated 160,440 deaths due to lung cancer.1 Seventy-five percent of patients with bronchogenic carcinoma will be diagnosed with non-small cell lung cancer (NSCLC). The number of patients with early or localized disease (currently an estimated 15-20% of NSCLC patients)2 is expected to rise over the next several years due to widespread screening with CT scanning. The treatment of choice for stage I (T1-T2N0) NSCLC is surgical resection which results in 5-year survival rates of approximately 60 to 70%.3-5 Occasionally, however, there are patients with early-stage NSCLC that are unable to tolerate surgical resection or the postoperative recovery period due to various comorbidities. While conventionally fractionated radiation therapy has been utilized as nonsurgical therapy for these medically inoperable patients, close observation with no specific cancer therapy has also been advocated in highly selected cases. McGarry, et. al., reviewed outcomes in 75 patients who had received no specific cancer therapy for stage I NSCLC, and the cause of death was progressive cancer in 53% of cases with a median survival time of 14.2 ± 2.4 months.6 Definitive conventionally fractionated RT for early-stage NSCLC is considered reasonable non-surgical therapy but yields poor 5-year survival rates ranging from 10 to 30%.7-11 Several studies have suggested a dose-response relationship reporting a benefit to dose escalation above the standard conventionally fractionated 4,500 to 6,600 cGy. This benefit was evident in both survival and local control in these patients.10-14 Sibley, et. al., reviewed 156 medically inoperable patients with stage I NSCLC treated with primary RT at Duke University between 1980 and 1995. They reported a 5-year, cause-specific survival rate of 32%. There was a trend toward improved survival in those patients achieving local control which approached significance for higher RT doses (p = 0.07).13 At this institution, we have published a series treating 56 patients with medically inoperable NSCLC with a median dose of 70 Gy using conformal radiotherapy techniques.15 Actuarial local control rates were 69% and 63% at two- and three years of follow up, respectively. These data serve as the estimate for statistical power calculations for this trial. Radiation fields have historically encompassed the primary tumor as well as the regional lymphatics in the ipsilateral hilum and mediastinum. This elective treatment was based on the identified risk of occult lymph node involvement ranging up to 20% in some surgical series.16 In recent years, elimination of elective nodal irradiation, which is potentially poorly tolerated in this population,17 has been validated by several retrospective studies permitting treatment of the primary tumor alone with limited fields.18-21 Slotman, et. al., in a study from the Netherlands, reported the use of limited "postage-stamp" fields to treat early stage lung cancer patients using hypofractionated RT (i.e., 4,800 cGy in 400-cGy fractions). Reported 3-year overall and disease-specific survival rates were 42% and 76%, respectively.20 Most of the aforementioned retrospective studies utilized radiotherapy equipment from the era of 1-D and 2-D treatment planning. Several limitations are evident from these older techniques, including target visualization, selection of beam directions, and computational algorithms describing deposited dose. Recent improvements in software, hardware, and computer processing speed have revolutionized the delivery of radiation doses appropriate for tumor cell killing. In this new era of three-dimensional treatment planning, more precise delivery methods are available allowing for dose escalation to the target volume without excessive dose being deposited in normal tissues. The RTOG has completed an extensive dose escalation study of conventionally fractionated three-dimensional conformal radiotherapy (3D-CRT) for NSCLC for stages I, II, and III disease as long as all detectable tumor can be encompassed by the radiation therapy fields including both primary tumor and regional lymph nodes.22 No mechanism for minimizing lung and tumor movements was utilized. One hundred and seventy-nine patients were treated with radiation doses escalated to as high as 90.3 Gy. Patients were stratified within each dose level according to the percentage of the total lung volume that received \>20 Gy with the treatment plan (V20). For patients receiving radiation alone or radiation following induction chemotherapy, data from RTOG 9311 established that the radiation dose could be safely escalated using 3D-CRT techniques to 83.8 Gy for patients with V20 values of \<25% and to 77.4 Gy for patients with V20 values between 25% and 36%, using fraction sizes of 2.15 Gy. Excess mortality was observed at 90.3 Gy with two dose-related deaths. The incidence of grade 3 or higher acute toxicity is less than 10%; however, grade 3 or higher late toxicity was approximately 15%. Stereotactic Body Radiation Therapy Stereotactic body radiation therapy (SBRT) is the delivery of high precision, biologically potent doses of radiation to tumors of the chest, abdomen, and pelvis. Implementing elements of 3D-CRT with stereotactic targeting, SBRT permits delivery of 3-4 high dose fractions totaling 48-60 Gy with good local control and low toxicity. A phase I dose escalation trial has been completed at Indiana University for treatment of medically inoperable patients with stage I NSCLC.26, 27 SBRT was administered with large doses per fraction in an extracranial stereotactic body frame, which includes a system for decreasing breathing motion. The starting dose was 8 Gy times 3 (24 Gy total), and fraction dose was escalated by 2 Gy per fraction for each cohort. The target lesion was outlined by a physician and designated as the gross tumor volume (GTV). An additional 0.5 cm in the axial plane and 1.0 cm in the longitudinal plane was added to the GTV to constitute the PTV based on validation measurements for this commercially available system.23, 28, 29 Typically, 7 to 10 non-coplanar beams were used to encompass the PTV. Dose was prescribed to the 80% isodose line. However, higher isodoses occurred within the center of the target mimicking the heterogeneous dose profile common to intracranial stereotactic radiosurgery. The treatment isocenter was identified with 3-D coordinates defined stereotactically and localized on verniers attached to the frame. No skin or bony landmarks were used to set the treatment isocenter; however, orthogonal port films were used on a daily basis for isocenter verification.30 Separate dose escalations were carried out independently for patients with T1 versus T2 small (≤ 5 cm) versus T2 large (5-7 cm) tumors at diagnosis. According to the Indiana University protocol guidelines, dose-limiting toxicity (DLT) was any grade 3 cardiac or pulmonary toxicity or any grade 4 toxicity attributed to the therapy. Thirty-seven patients were treated using a standard dose escalation protocol with 3 patient cohorts with minimum 1 month between dose levels to assess toxicity. Patients were categorized into separate independent dose escalations according to tumor volume, T1 vs. T2 (≤ 5 cm) vs. T2 (\> 5 to ≤ 7 cm). Grade 3 pneumonitis was seen at a dose of 14 x 3 = 42 Gy total in one T2 patient with a 7-cm tumor and transient grade 3 hypoxia was seen at 16 x 3 = 48 Gy total in one patient. Additional patients were treated at each of these levels without further toxicity observed. Twenty-one patients had mild to moderate fibrosis distal to the treated lesion appear on chest x-ray after treatment. Nine of these patients had a decline of an element of their pulmonary function tests (FEV1, FVC, DLCO, or PO2) by 10-20% of predicted which returned back to baseline values with follow-up in all but two. The timing of onset of this toxicity was generally acute to subacute (\< 1 month in most cases). The maximum tolerated dose (MTD) was not reached on this trial for patients with T1 tumors and smaller T2 tumors (≤ 5 cm). Dose-limiting pneumonitis or pericarditis occurred in 2/5 patients with larger T2 tumors (\>5 to ≤ 7 cm) at a dose of 24 x 3 = 72 Gy defining the MTD for this subgroup at 22 x 3 = 66 Gy. Patients treated at a dose of 22 Gy per fraction times three fractions had follow up of over 24 months without late toxicity for all T-stage tumor categories. Treatment failure within the PTV has been observed in 8 of 26 patients treated at doses of up to 20 x 3 = 60 Gy. However, all but one of these local failures occurred at doses of 16 x 3 = 48 Gy or lower.31 The above data demonstrate that solitary lung lesions including early stage NSCLC are better controlled with SBRT when compared to conventional radiation. In addition, reduced volume treatments are attractive in these patients with medically inoperable stage I NSCLC who may have an increased risk of radiation pneumonitis associated with conventional large volume radiation fields. SBRT permits dose escalation by significantly reducing the high-dose treatment volume. The RTOG opened in May 2004 a phase II trial of SBRT in the treatment of medically inoperable patients with stage I/II non-small cell lung cancer in an effort to determine if SBRT could achieve acceptable local control as seen in retrospective series.24, 26, 32-38 A secondary objective is to determine if this technique achieves acceptable treatment-related toxicity. In this trial, patients with T1, T2 (≤ 5 cm), or T3 (≤ 5 cm), N0, M0 medically inoperable non-small cell lung cancer are treated with SBRT to a total of 60 Gy in 3 fractions of 20 Gy each over 1.5 to 2 weeks. This protocol excludes patients with T3 tumors involving the central chest and structures of the mediastinum as well as patients with any T-stage tumor within or touching the zone of the proximal bronchial tree. This region is defined as a volume 2 cm in all directions around the proximal bronchial tree (carina, right and left main bronchi, right and left upper lobe bronchi, intermedius bronchus, right middle lobe bronchus, lingular bronchus, and right and left lower lobe bronchi).