Radiobiological Determination of Dose Escalation and Normal Tissue Toxicity in Definitive Chemoradiation Therapy for Esophageal Cancer()

PURPOSE: This study investigated the trade-off in tumor coverage and organ-at-risk sparing when applying dose escalation for concurrent chemoradiation therapy (CRT) of mid-esophageal cancer, using radiobiological modeling to estimate local control and normal tissue toxicity. METHODS AND MATERIALS: Twenty-one patients with mid-esophageal cancer were selected from the SCOPE1 database (International Standard Randomised Controlled Trials number 47718479), with a mean planning target volume (PTV) of 327 cm(3). A boost volume, PTV2 (GTV + 0.5 cm margin), was created. Radiobiological modeling of tumor control probability (TCP) estimated the dose required for a clinically significant (+20%) increase in local control as 62.5 Gy/25 fractions. A RapidArc (RA) plan with a simultaneously integrated boost (SIB) to PTV2 (RA(62.5)) was compared to a standard dose plan of 50 Gy/25 fractions (RA(50)). Dose-volume metrics and estimates of normal tissue complication probability (NTCP) for heart and lungs were compared. RESULTS: Clinically acceptable dose escalation was feasible for 16 of 21 patients, with significant gains (>18%) in tumor control from 38.2% (RA(50)) to 56.3% (RA(62.5)), and only a small increase in predicted toxicity: median heart NTCP 4.4% (RA(50)) versus 5.6% (RA(62.5)) P<.001 and median lung NTCP 6.5% (RA(50)) versus 7.5% (RA(62.5)) P<.001. CONCLUSIONS: Dose escalation to the GTV to improve local control is possible when overlap between PTV and organ-at-risk (<8% heart volume and <2.5% lung volume overlap for this study) generates only negligible increase in lung or heart toxicity. These predictions from radiobiological modeling should be tested in future clinical trials.