Prognostic factors in ambulatory patients with inoperable locoregionally recurrent rectal canter following curative surgery

Background: The optimal treatment for locoregionally recurrent rectal cance r after curative surgery has not yet been defined The definition of prognos tic factors could lead to the selection of an aggressive therapeutic approa ch in patients with favourable prognosis alone. Patients and methods: The r ecords of thirty-nine ambulatory pts, 15 female and 24 male, with diagnosis of locoregionally recurrent rectal cancer (LRRC) after curative surgery an d treated with radiotherapy were retrospectively analyzed. The following fa ctors were analyzed for their ability to predict the clinical response and outcome for LRRC: age, sex, initial tumor grading, primary surgical approac h, initial primary tumor stage according to Dukes' classification, disease free survival (time to primary surgery and detection bf a LRRC), pelvic-per ineal structure affected by recurrence, total radiation dose, chemotherapy with fluorouracil, symptomatic response to the therapy, locoregional sympto matic re-recurrence, systemic progression disease. Results: In the univaria te analysis, predictive factors for survival, were graded (G1-2 vs G3 p=0.0 4), Dukes' stage at first diagnosis (A-B vs C p=0.01), and site of pelvic-p erineal recurrence (Pelvic mass alone yes vs no p=0.01; Nerve and/or Osseou s involvement yes vs no p<0.001). Following therapy for LRRC, a better surv ival was observed in pts with a complete symptomatic response (complete rem ission vs partial remission vs no change p<0.001), without a further locore gional symptomatic re-recurrence (re-recurrence yes vs no p=0.001) and/or a ppearance of metastatic disease (yes vs no p<0.001). Multivariate analysis showed Dukes' stage (p=0.01, RR=0.5, cr 0.3-0.8), Grading (p=0.01, RR=0.5, CI 0.3-0.9), initial Surgical approach (p=0.02, PR=1.9, CI 1.1-3.3) and Pel vic mass alone (p=0.001, RR=0.4, CI 0.2-0.7) to be statistically significan t. Conclusion: Good performance status, Grading 1-2, Dukes' stage A-B at fi rst diagnosis of rectal cancel; initial conservative surgical approach and LRRC without pelvic nerve and/or osseous involvement indicate a favourable prognosis in LRRC. The results indicate the definition of a subgroup of pat ients for whom an aggressive therapeutic approach could be justified (exter nal radiotherapy surgery and intraoperative radiation therapy).