2101 - Incidence and Dosimetric Correlates of Radiation-Induced Lung Injury after Adjuvant Radiotherapy for Gastric Cancer

03:00pm - 04:00pm ET

Poster Hall - Exhibit Hall A

Screen: 27

Presenter(s)

Hatice Kubra Isik, MD - Marmara University Pendik Education and Research Hospital Radiation Oncology Clinic, Istanbul, Marmara

H. K. Isik¹, Ö. Demircioglu², M. Buyukkomurcu², T. Tasçi³, and H. Alkis¹; ¹Department of Radiation Oncology, Marmara University School of Medicine, Istanbul, Turkey, ²Marmara University Department of Radiology, istanbul, Turkey, ³Marmara University Pendik Education and Research Hospital Radiation Oncology Clinic, Istanbul, Turkey

Purpose/Objective(s):

Radiation induced lung injury (RILI) represents a significant complication predominantly associated with irradiation of thoracic malignancies. However, RILI may also emerge with radiotherapy (RT) of non-thoracic malignancies if the radiation field encompasses lungs. It is often underestimated or even disregarded in gastric cancer patients because of the relatively low pulmonary dose exposure. In this study, we aimed not only to evaluate the incidence of RILI in patients receiving adjuvant RT for gastric cancer, but also to demonstrate that even low-dose, out-of-field lung exposure is not biologically inert.

Materials/Methods:

Patients with gastric cancer who received adjuvant RT were included. Thoracic computed tomography (CT) scans obtained within 3–6 months after completion of RT were fused with pretreatment simulation CT. RILI areas were contoured by two experienced radiologists. Patients were categorized as RILI-positive and RILI-negative. Dosimetric parameters for total, right and left lung were documented. Out-of-field lung regions with RILI areas were classified according to very low-dose exposure as <5 Gy and =5 Gy. Mann-Whitney U test was used for statistical analysis.

Results:

A total of 33 patients were included. Median age was 59 (30–84), and female/male was 7/26. Median RT dose was 45 Gy (45-50.4). All patients received concurrent chemotherapy. While RILI was detected in 16 patients, 17 patients did not have radiological finding of RILI. In RILI positive patients total lung mean dose (Dmean) was significantly higher compared to RILI negative patients(5.28 ± 3.04 vs 3.46 ± 1.1, p=0.036). A total of 66 RILI areas were detected (36 in right and 30 in left lung). Median dose of the RILI areas was 2.75 Gy (0.25-45), 4 Gy (0.25-45), and 2.25 Gy (0.25-45) for total, right, and left lungs, respectively. RILI areas were with very low dose regions (<5 Gy) in 12 patients. Of the 66 RILI areas, 38 were in <5 Gy and 28 in =5 Gy regions (mean dose: 1.43 ± 0.9 Gy vs 20.25 ± 12.34 Gy, p<0.0001). Mean percentage dose of RILI areas was significantly higher in the regions receiving <5 Gy than =5 Gy (49.47 ± 29.9 vs 13.84 ± 13.34, p<0.0001). In the left lung, 30 RILI areas were detected in 15 patients; of them, 20 were in regions receiving <5 Gy and 10 in =5 Gy (1.66 ± 1.06 Gy vs 28.1 ± 13.48 Gy, p<0.0001). In the right lung, 36 RILI areas were identified in 11 patients; 18 were in regions <5 Gy and 18 in =5 Gy (1.17 ± 0.8 Gy vs 15.89 ± 9.44 Gy, p<0.0001).

Conclusion:

Greater number of RILI areas in low-dose and out-of-field regions of the right lung suggest that efforts to minimize incidental irradiation of uninvolved lung tissue may be clinically relevant. Even low dose out-of-field lung exposure is not biologically inert even when clinically silent subclinical radiation injury. Furthermore, the potential contribution of the abscopal effect of RT and factors such as COPD and immunologic disorders should be considered to minimize incidence of RILI.