Sargent, Daniel J.

656 Background: Patients (pts) with peritoneal metastases from CRC (pmCRC) have reduced OS compared to mCRC pts without peritoneal involvement. Here we further investigated the impact of number and location of metastases among pts receiving first-line systemic chemotherapy. Methods: Individual patient data were available on 9,265 pts (median age 64; 63% male; 93% ECOG PS 0-1; 68% colon primary tumor; brain metastases excluded) enrolled onto 12 first-line randomized trials (4 tested targeted regimens). Stratified multivariable Cox models were used to assess the associations with overall survival (OS); adjusted hazard ratios (HRadj) and 95% confidence intervals are reported (CI). Results: There were 7,963 (86%) pts with non-pmCRC (3,904 with one disease site; 4,059 with ≥2 disease sites), 191 (2%) pts with isolated pmCRC, and 1,111 (12%) non-isolated pmCRC. These groups were similar in age, race, and use of targeted chemotherapy. Compared to non-pmCRC, pts with pmCRC were more likely to be female (41% vs. 36%, p<.001), have colon primary tumors (85% vs. 67%, p<.0001), and have PS2 (10% vs. 6%, p<.0001). Compared to isolated pmCRC, pts with solitary non-peritoneal sites (both M1a) had significantly better OS (HRadj=0.78; CI, 0.64-0.94, p=.009) while pts with ≥2 non-peritoneal sites had similar OS (HRadj=1.06; CI 0.88-1.28, p=.535). OS of pts with pmCRC with a single other disease site (n=446) was similar to isolated pmCRC (HRadj=1.13; CI 0.91-1.40, p=.28), but those with pmCRC + ≥2 additional disease sites (n=665) had shortest survival (HRadj=1.44; CI 1.17-1.77, p<.001). A combination of peritoneal and liver metastases (n=821; HRadj=1.37, CI 1.12-1.67, p=.002) was associated with poorer survival compared with isolated pmCRC; but combination with extrahepatic sites (n=290; HRadj=1.15, CI 0.91-1.45, p=.25) was not. Conclusions: pmCRC pts have significantly worse survival than those with other solitary site mCRC. Among those with multiple disease sites, poorer survival is a function of increased number of metastatic sites and peritoneal involvement, which indicates prognostic heterogeneities among M1b pts.

PURPOSE: The toxicity profile of fluorouracil (5-FU)–based chemotherapy given on 5 consecutive days at doses of 370 to 450 mg/m2 has been well documented. A meta-analysis of six North Central Cancer Treatment Group (NCCTG) cancer control trials involving 786 patients indicated that women treated with this type of regimen experienced more severe stomatitis and leukopenia than men. After these findings, an additional meta-analysis of the toxicity profiles on five NCCTG colorectal cancer treatment trials was undertaken. METHODS: Data for 1,093 women and 1,355 men from 12 different treatment arms were included. The primary end points were the incidence of stomatitis, leukopenia, alopecia, diarrhea, nausea, and vomiting, recorded with standard National Cancer Institute common toxicity criteria. Fisher’s exact test was used to compare incidence and severity across sexes, supplemented by Forrest meta-analysis plots and logistic regression. RESULTS: The incidence of four out of six toxicities studied was significantly greater for women than men; the exceptions were severe nausea and vomiting. Overall, almost half of the women compared with a third of the men experienced severe toxicity (P < .0001). Logistic regression confirmed the univariate findings while adjusting for the effects of study, dose, body mass index, and age. The differences were consistent across treatment cycles. Response rates and survival distributions were the same for both sexes. CONCLUSION: This study confirms an earlier finding that women receiving 5-FU–based chemotherapy in a 5-day bolus schedule experience toxicity more frequently and with more severity than men. These data raise the question of whether the recommended initial dose of 5-FU–based chemotherapy for women should be lower than that for men.

PURPOSE: Substantial and successful effort has been focused on decreasing the risk of local failure after rectal cancer surgery through the use of adjuvant therapies. Our study examined data from studies conducted by United States cooperative groups to investigate the impact of surgical and pathologic variables in rectal cancer outcomes. PATIENTS AND METHODS: Surgical and pathologic reports from 673 patients with stage II/III rectal cancer enrolled onto three adjuvant clinical trials were reviewed for tumor and surgical variables. Additional information on individual institutions and operating surgeon was collected. Variables were tested for association with 5-year local recurrence and survival after adjustment for adjuvant treatments and other important prognostic factors. RESULTS: Five-year local recurrence and survival rates were 16% and 59%, respectively. Surgeons treating more than 10 study cases had lower local recurrence rates than those treating ≤ 10 (11% v 17%, P = .02). Free radial margins also correlated with local recurrence (P = .01). Type of surgery, distal margins, and tumor radial spread were not significant. Tumor adherence to adjacent structures predicted local recurrence (35% v 14%, P < .001) and survival (30% v 63%, P < .001), regardless of en bloc resection. Although T and N classification predicted survival (P < .001), only N classification correlated with local recurrence. The number and percentage of positive nodes correlated with survival, but only the percentage independently predicted local recurrence. Several pathologic and surgical variables were reported suboptimally. CONCLUSION: Moderate variability in outcomes among surgeons was detected in this high-risk population. Efforts to improve surgical results will require changes in reporting practices to allow for more accurate assessment of the quality of surgery.

659 Background: Prospective survival prediction of patients with metastatic colorectal cancer is difficult. Prognosis estimation based on readily available clinicopathologic factors has the potential to inform clinical practice and improve risk stratification for clinical trials. We constructed prognostic nomograms for OS and PFS in mCRC using the multi-trial ARCAD database. Methods: Data from 19,678 mCRC pts accrued to 24 first line randomized phase III clinical trials since 1997 were used to construct and validate Cox models for PFS and OS, stratified by treatment arm within each study. Candidate variables included age, gender, BMI, performance status, colon vs. rectal cancer, prior chemotherapy, number of metastatic sites, sites of metastases (liver, lung, lymph nodes), and baseline bilirubin, albumin, white blood cell count, hemoglobin, platelets, absolute neutrophil count, and derived neutrophil:lymphocyte ratio (dNLR). Missing data (<11%) were imputed, continuous variables modeled with splines, and clinically relevant pairwise interactions considered if p<0.001. Final models were internally validated via bootstrapping to obtain optimism-corrected calibration and discrimination C-indices, and externally validated using a 10% holdout sample from each trial. Results: Nomograms for OS and PFS including remaining variables were well calibrated with C-indices of 0.66 and 0.60, respectively. Evaluation of external validity revealed good concordance; 71% and 67% respectively between predicted (> vs. <50% probability) and actual (yes/no) 1-year OS and 6-month PFS, and median 1-year OS and 6-month PFS predictions fell within the actual 95% Kaplan-Meier intervals. Gender, liver and lung metastases, and dNLR were not prognostic for OS; prior chemo, colon vs. rectum, dNLR, liver and lymph node metastases, and gender did not predict for PFS. No clinically relevant pairwise interactions were identified. Conclusions: The proposed nomograms are well calibrated and internally and externally valid. These tools have the potential to aid prognostication and patient/physician communication, and balance risk in randomized trials in mCRC.