Impact of Target Volume Margins on Locoregional Control and Acute Toxicity for Unresectable Non-Small Cell Lung Cancer

Abstract

Purpose/Objective(s): Modalities such as image-guided radiation therapy and better understanding of tumor motion have enhanced tumor targeting, enabling a decrease in target volume margins. With newer imaging/treatment technologies, it is unclear the extent to which clinicians can reduce standard margins for inoperable non-small cell lung cancer (NSCLC), especially in the setting of concurrent chemotherapy which should treat microscopic disease. We hypothesized that smaller target volume margins, when used in conjunction with 4-dimensional computed tomography (4DCT) simulation and daily cone beam computed tomography (CBCT), would not compromise locoregional control and would decrease acute toxicity. Materials/Methods: We conducted a retrospective review of 82 patients with Stage II-III (n Z 73) or Stage IV oligometastatic (n Z 9) NSCLC who received definitive and concurrent chemoradiation therapy (CRT) and underwent 4DCT simulation and daily CBCT. Real time fluoroscopy with abdominal compression was used before delineation of an internal target volume (ITV) on the 4DCT, which was further expanded to a clinical target volume and planning target volume (PTV). ITV to PTV margins were typically 6 mm-10 mm. As another way to quantify variations in margins, a standard PTV was generated by expanding the original gross tumor volume (GTV) with a 10 mm margin, and a margin deviation index (MDI) was calculated as the percentage change in equivalent uniform dose between the original PTV and standard PTV. Grade 3 radiation esophagitis (RE3) and pneumonitis (RP3) were assessed with CTCAE v4.03. Spearman rank correlation tests, the cumulative incidence method, Gray's test, and Fine-Gray method were used. Results: Median MDI was 2.1% (range,-4.0-55.7%), with MDI 5.8% for 33% (n Z 27/82) of patients. Median ITV to PTV margin was 8 mm (range, 0-15 mm). A greater MDI correlated with a smaller ITV to PTV margin (p < 0.001), a larger GTV (p < 0.001), and use of intensitymodulated radiation therapy (IMRT) (p Z 0.002). A tighter target volume margin (greater MDI and smaller ITV to PTV margin) did not increase the risk of locoregional progression (MDI 2.1% vs < 2.1%, HR 0.59, p Z 0.15; MDI 5.8% vs < 5.8%, HR 0.82, p Z 0.62; ITV to PTV margin < 8 mm vs 8 mm, HR 0.61, p Z 0.21). MDI 5.8% vs < 5.8% was associated with a marginal decrease in 1-year cumulative incidences of RE3 (4% vs 24%, p Z 0.025) and either RE3 or RP3 (11% vs 31%, p Z 0.043), while MDI 2.1% vs < 2.1%, ITV to PTV margin < 8 mm vs 8 mm, and IMRT were not significantly associated with these toxicities. Conclusion: Smaller target volume margins were used for larger tumors, yet did not increase the risk of locoregional progression. Additional studies are needed to determine if smaller margins, when used alongside 4DCT simulation and daily CBCT, can be employed consistently to decrease toxicity during definitive CRT for inoperable NSCLC.