Usefulness of CEA, TPA, GICA, CA 72.4, and CA 195 in the Diagnosis of primary colorectal cancer and at its relapse.

A protocol with tumor markers as guidelines to follow up colorectal cancer patients was designed using criteria other than those commonly reported. They included combination of several markers and their dynamic evaluation of three different levels of increase: isolated elevated value (IEV), constant level of elevation (CE), and progressive increase (PI). In a total of 90 patients, the levels of combined serum CEA-TPA and GICA were serially measured, and in 71 of them, CA 72.4 and CA 195 levels were also determined. The tumor markers were measured during the first few months after surgery, and the usefulness of combined CEA-TPA-GICA and other, possibly more favorable combinations was determined in relation to "early" detection of recurrence and development of metastases. In addition the usefulness of conventional radiologic examinations and the impact on patients survival following "early" diagnosis was evaluated. A positive correlation was found between elevated preoperative serum tumor marker levels and the stage of disease. The postoperative variation of high serum CEA values was useful in identifying micrometastases after primary tumor resection. In the "early" diagnosed 14 patients with recurrence during the postoperative follow-up period, the highest sensitivity was found for TPA (87%) and, of the marker combinations, TPA-GICA (93%) with a lead time of 4.6 +/- 5.6 and 5.4 +/- 7.8 months (mean +/- SD) respectively. In nonrelapsed patients, falsely positive results of TPA-GICA (25%) were fewer than those for TPA-CA 195 (31%) and TPA-GICA-CA72-4 (35%). However, TPA-CA 195 and TPA-GICA-CA72-4, based upon their high sensitivity in patients with metastases, seemed in keeping with the effectiveness of TPA-GICA for monitoring of postoperative patients with colorectal cancer. In patients who developed recurrences, PI was more frequently present than IEV. In patients without recurrence, the opposite occurred. CE had less frequently discriminatory capability between these two groups than IEV and PI. Routine radiographic studies were ineffective whereas liver echography with its high sensitivity revealed the first sign of recurrence. Eight (50%) of the 16 relapses (two patients relapsed twice) were suitable for surgery because only one organ with a single metastasis was involved. Three (75%) of the 4 patients with "early" diagnosis of recurrence are alive without evidence of disease 5, 18, and 20 months after the last surgery. The results of this study revealed the importance of "early" diagnosis of recurrence for improved survival of patients with colorectal cancer.