LATE EFFECT OF MULTIPLE DAILY FRACTION PALLIATION SCHEDULE FOR ADVANCED PELVIC MALIGNANCIES (RTOG-8502)

Purpose: Determine late complication incidence for pelvic palliation u sing accelerated multiple daily fraction radiation [Radiation Therapy Oncology Group (RTOG) 8502]. Methods and Materials: Prospective evalua tion of a palliative radiation schedule for advanced pelvic malignanci es was conducted from 1985 to 1989 by RTOG 8502. The dose was 44.40 Gy in 12 fractions (3.7 Gy BID) with a rest after 14.80 Gy and 29.60 Gy. The pilot part of the study allowed for a variable rest interval of 3 -6 weeks. The rest interval was then randomized between 2 and 4 weeks to determine effect on tumor control. No difference in tumor control w as identified (p = 0.59). The pilot segment accrued 151 patients and t he randomized segment accrued 144 patients. A total of 290 cases were analyzable (four ineligible or canceled) for late effects. To minimize actuarial bias, only patients surviving 90 days (193) were analyzed f or late effect risk. The primary site consisted of gynecologic (40%), colorectal (28%), genitourinary (25%), and miscellaneous (7%). The ext ent of tumor consisted of pelvis only (62%) and additional tumor outsi de the pelvis (38%). Most of the patients were elderly (76% > 60 years , 47% > 70 years). Karnofsky performance status (KPS) was greater than or equal to 80 in 60% of patients and < 80 in 40%. Results: None of t he patients with < 30 Gy (less than three courses) developed late toxi city. A total of 11/193 (6%) developed Grade 3+ late toxicity (nine Gr ade 3, one Grade 4, one Grade 5). Actuarial analysis of complication r ate by survival time for Grades 3, 4, and 5 shows a cumulative inciden ce for complications after 6 months that plateaus at 6.9% by 18 months . The cumulative incidence for Grades 4 and 5 is 2.0% by 12 months. Th e difference in late effect for the 2-week rest vs. 4-week rest was no t statistically different (p =.47). Patient factors evaluated for incr eased risk of late complications included prior surgeries, age, sex, K PS and primary. None were found to have significant statistical correl ations with late effects. Conclusion: The crude late complications rat e is 6%. Actuarial analysis using cumulative incidence shows 6.9% by 1 8 months. This represents a significant decrease in late complications from 49% seen with higher dose per fraction (10 Gy X 3) piloted by Ra diation Therapy Oncology Group (7905) for a similar group of patients. Long-term analysis of late complication indicates this schedule can b e used in the pelvis with relatively low incidence of complication. Th is schedule has significant logistic benefits and has been shown to pr oduce good tumor regression and excellent palliation of symptoms.