FCCC LOGO Faculty Publications
Katz A , Hanlon A , Lanciano R , Hoffman J , Coia L
Prognostic value of CA 19-9 levels in patients with carcinoma of the pancreas treated with radiotherapy
International Journal of Radiation Oncology Biology Physics. 1998 May 1;41(2) :393-396
PMID: ISI:000073743100021   
Back to previous list
Purpose: CA 19-9 has been identified as a tumor marker for pancreatic carcinoma and has been shown to have some utility in predicting outcome in surgically treated patients. The purpose of this study was to evaluate its usefulness as a prognostic indicator in patients treated with radiotherapy. Materials and Methods: A retrospective review of all patients treated with radiotherapy of definitive intent (n = 104) for carcinoma of the pancreas at Fox Chase Cancer Center from 1980-1994 was undertaken. Patients were categorized into four groups: Planned preoperative radiation with resection (n = 25); planned preoperative radiation without successful resection (n = 35); postoperative radiation (n = 21); and radiation without planned resection (n = 23). For each group, except those treated without planned resection, median dose for external beam radiotherapy was 50.4 Gy (range = 21.6-63.0 Gy). Those in the remaining fourth group were treated with a median dose of 55.8 Gy (range = 36.0-60.4 Gy). 97% of patients in the first three groups were also treated with 5-FU-based chemotherapy, as were 61% of those in the fourth group. Pretreatment and follow-up CA 19-9 levels were available for 69 patients. Results: Median survival time for all groups was 10 months (range = 1-67 months). Univariate analysis showed significant differences in survival among the groups: Preop with resection 22 months; preop without resection 10 months, postop 17 months; and without planned resection 12 months (p = 0.0005). Overall, patients who underwent resection had a median survival time of 19 months, compared to 11 months in those who did not (p = 0.0006). CA 19-9 level at diagnosis was found to be a significant prognostic indicator on univariate analysis, with a median survival time of 8 months in those having a level greater than the median! of 680 U/ml, compared to 20 months in those who did not (p = 0.0003). Similarly, the posttreatment nadir was significant, with a median survival time of 11 months in those with levels above the median of 162.5 U/ml, vs. 26 months in those with levels below 162.5 U/ml (p = 0.001). The median survival time for patients whose CA 19-9 levels decreased in response to treatment by more than 75% was 23 months (range = 6-34 months) vs. 8 months (range = 3-21) in those with 75% or less response (p = 0.003). On stepwise multivariate analysis, pretreatment CA 19-9 level was found to be a significant predictor of survival (p = 0.005). Other potential indicators of outcome, including age, gender, KPS, prediagnosis weight loss, location of tumor, clinical TNM staging, size of lesion, vascular involvement on angiography, and sequence of radiation with respect to resection, were evaluated and were not found to be significant. Conclusion: CA 19-9 was demonstrated to be a useful prognostic indicator in patients treated with radiotherapy; other, more traditional, indicators of outcome were of less utility. (C) 1998 Elsevier Science Inc.
Times Cited: 6 English Article ZP346 INT J RADIAT ONCOL BIOL PHYS