Outcome After Stereotactic Radiotherapy in 'High-Risk' Patients With Stage I Non-small Cell Lung Cancer (NSCLC)
Reviewer: Christine Hill, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 31, 2007
Presenter: F. J. Lagerwaard Presenter's Affiliation: VU University Medical Center, Amsterdam, The Netherlands Type of Session: Scientific
Surgical treatment with lobectomy or pneumonectomy is recognized as the optimal treatment for stage I non-small cell lung cancer (NSCLC) (Kraev A, Chest, 2007); however, many patients with this disease are not good surgical candidates due to comorbidities, some of which may be associated with prolonged tobacco use.
Stereotactic radiotherapy (SRT) has emerged as a potential treatment for patients with stage I NSCLC who are not surgical candidates or who choose to avoid surgery (Onishi, J Thoracic Oncol, 2007). The use of this treatment is well-described for patients with tumors of T1 classification.
Some reports have suggested increased toxicity when SRT is used for patients with larger, T2, tumors or tumors located centrally near the hilum, mediastinum, heart, or spinal cord.
Patients at the VU University Medical Center have been treated with SRT for stage I NSCLC since 2003. This report is an analysis of the complications encountered by patients with either T2 tumors, or T1-T2 tumors that are centrally located.
Materials and Methods
247 patients treated with SRT for stage I NSCLC were analyzed within this study. Mean follow-up was 20 months.
80% of the patients considered in this study were deemed inoperable, and the remaining 20% refused surgery.
Patients were considered “low-risk” if their tumors were T1 and peripherally located. Tumor size ≥ 3 cm, or presence of any size tumor within 2 cm of the hilar or mediastinal structures were criteria for classifying a patient as “high-risk.”
Pathologic confirmation of malignancy was available for 38% of patients. Other patients had new or growing PET-positive lesions with CT characteristics of malignancy.
Internal target volumes (ITV) were generated using 4-dimensional CT scans, and an isotropic 3 mm margin was added for delineation of the planning target volume (PTV).
SRT was delivered with 8 – 12 non-coplanar fields with micro-multileaf shielding using the ExacTrac-system (BrainLab, Feldkirchen, Germany). Fractionation schemes were as follows, and all doses were prescribed at the 80% isodose line:
For patients with peripheral T1 tumors, 3 fractions of 20 Gray (Gy) were delivered, for a biologically equivalent dose (BED) of 180 Gy.
For patients with peripheral T2 tumors, 5 fractons of 12 Gy were delivered (BED = 132 Gy).
For patients with centrally located tumors, 8 fractions of 7.5 Gy were delivered (BED = 105 Gy).
No patient in this study was treated with chemotherapy.
Of the 247 patients considered in this study, 127 had “low-risk” tumors, and 127 had “high-risk” tumors. Of the tumors considered high risk, 19 were centrally-located T1 tumors, 30 were centrally-located T2 tumors, 54 were peripherally-located T2 tumors, and 17 were multiple.
Of the high risk patients, 75% were treated with the 5 fractionation scheme, and 25% were treated with the 8 fractionation scheme. Of low-risk patients, 76% were treated with 3 fractions, and 24% were treated with 5.
Pathologic tumor evaluation was available for 43% of high-risk patients and 27% of low-risk.
Mean overall survival (OS) was 34 months for high-risk patients and 36 months for low-risk patients (p = 0.38).
Disease-free survival rates (DFS) at one year were 59% for high-risk patients and 76% for low-risk (p = 0.017).
Local failure rates were 12% for high risk patients and 3% for low-risk (p = 0.13), and rates of distant metastatic failure were 32% for high risk patients and 19% for low-risk (p = 0.004).
On further analysis, tumor stage, FEV1, and performance status were prognostic factors for OS in patients with high risk disease. Tumor diameter was noted to be of prognostic significance only for DFS in high-risk patients considered in this study. Age was the only factor of prognostic significance for low-risk patients.
No overall difference in outcomes was noted based on fractionation scheme for either group of patients.
Of high-risk patients, 50% experienced no significant toxicity. The most common symptoms experienced, by 11% of high-risk patients, were fatigue and nausea. Other toxicities correlated positively with increasing PTV size, and included clinically significant pneumonitis in 3%, chronic pain in 3%, and rib fracture in 2% of high-risk patients.
The authors conclude that SRT based upon 4-dimensional treatment planning is a feasible and safe treatment regimen for patients with centrally located or classification T1 tumors in the setting of stage I NSCLC.
They conclude that the treatment is effective when a BED ≥ 100 Gy is used, resulting in reasonable rates of local control.
They note that, despite pre-treatment PET scanning, distant failure remains the most common pattern of relapsed/ recurrent disease.
This study represents an analysis of a relatively large cohort of patients with tumors that might be considered inappropriate for SRT treated with this modality.
The rates of acute toxicity were acceptable, as the authors note; however, one of the concerns with use of SRT for larger or centrally located tumors is late toxicity. Use of hypofractionated radiotherapy is known to result in increased late normal tissue damage, much of which will not be evidenced for 10 – 20 years after completion of treatment. The limited median follow-up of 20 months in this study does not allow for assessment of potential late effects resulting from hypofractionated, high radiation doses given to larger volumes of tissue or vital organs such as the heart, esophagus, and spinal cord. Only time will allow full assessment of these effects; however, particular attention should be paid to volume of organs at risk receiving hypofractionated radiation during treatment such as this. Also, because this is a selected patient population with significant comorbidities, the patients may not live long enough to demonstrate late toxicity.
Given the relatively high risk of distant metastatic failure in this patient population, consideration of chemotherapy for patients with larger tumors could be considered in the future. In fact, benefit of chemotherapy added to surgery has been demonstrated for patients with stage IB NSCLC by the Cancer and Leukemia Group B (CALGB) and the National Cancer Institue of Canada (NCIC), and these results would be expected to extrapolate to apply to patients treated non-surgically for stage IB disease. Further investigations in this vein would be indicated and appropriate.
The low rate of pathologically proven NSCLC in the population of patients considered is somewhat concerning, as over half of the patients considered were in fact not proven to have NSCLC. Based on epidemiologic considerations, at least some of them would be expected to have in fact other diagnoses of malignancy (such as SCLC or metastatic disease from a separate primary site), or even non-malignant diseases (such as histoplasmosis).
In this study, patients with “high-risk” stage IB NSCLC were treated with SRT with appropriate acute toxicity and local control rates. Further investigation with regards to late toxicity and addition of chemotherapy are certainly warranted; in the meantime, use of hypofractionated, high-dose radiotherapy in close proximity to vital organs at risk should be undertaken cautiously.
Partially funded by an unrestricted educational grant from Bristol-Myers Squibb.