BackgroundA prospective phase II study was performed to determine the feasibility, efficacy and safety of arterial hepatic infusion (HAI) using pirarubicin combined with intravenous chemotherapy.Patients and methodsFrom December 1991 to April 1994, 75 patients with unresectable colorectal metastases confined to the liver were included in this multicenter study to receive intra-arterial hepatic pirarubicin and a systemic monthly regimen of 5-fluorouracil (5-FU) and folinic acid. Sixty-four patients were analyzed in the intention-to-treat analysis and 61 in the per-protocol analysis.ResultsTolerance of this regimen was rather good; however, functional catheter problems were observed in 29 patients (45%) resulting in failure of HAI in 21 cases (33%) after a median of three cycles; vomiting grade 3 was present in 12.5% of patients, neutropenia grade 4 in 23% and alopecia grade 3 in 19%. The overall response rate was 31.9% in intention-to-treat analysis, and 39.3% in per-protocol analysis. Extrahepatic progression was reported in only 21.7% of patients. Time to hepatic progression and extra-hepatic progression was 8.3 and 15 months, respectively, in intention-to-treat analysis, and 11 and 18 months, respectively, in per-protocol analysis. Median survival was 19 and 20 months in intention-to-treat analysis and per-protocol, respectively.ConclusionsIn our study, the combination of intra-arterial pirarubicin and intravenous chemotherapy demonstrated some efficacy and good tolerance in the treatment of isolated colorectal liver metastases. This treatment seems to prevent extra-hepatic spread and prolong survival time. The results of this study have to be confirmed by new trials using more active systemic chemotherapy.

The aim of our study was to assess the relationship between colorectal tumor responsiveness to irinotecan and microsatellite instability (MSI), a feature of colorectal tumors with DNA mismatch repair defect. Seventy-two patients with metastatic colorectal cancer were included in our retrospective study. A complete response to irinotecan was observed in 1 patient and a partial response in 10 patients, whereas 61 patients did not respond to this treatment. We analyzed the protein expression of hMLH1, hMSH2, and BAX by immunohistochemistry, determined the MSI phenotype, and looked for mutations in the coding repeats located in the transforming growth factor beta-RII, BAX, hMSH3, and hMSH6 genes. All 44 tumors analyzed expressed detectable levels of hMLH1; 1 tumor lacked hMSH2 staining, whereas 4 tumors showed a marked decrease in BAX expression. A better response to irinotecan was observed in the patients whose tumors have lost BAX expression (P < 0.001). Among the 7 tumors that displayed a MSI-H phenotype, 4 responded to irinotecan, whereas only 7 of the 65 MSI-L/ microsatellite stable tumors did (P = 0.009). Seven of the 72 tumors had inactivating mutations in the coding repeats of the target genes. Three tumors displayed a mutation in the poly-A10 tract of the transforming growth factor beta-RII gene, associated with a 1-bp deletion in the poly-A8 tract of hMSH3 in one tumor and with a 1-bp deletion in the poly-G8 tract of BAX in another. Four tumors displayed mutations in the poly-G8 repeat of BAX, whereas 2 mutations in hMSH6 and hMSH3 were characterized. Among the 7 tumors with mutations in these target genes, 5 responded to irinotecan, whereas only 6 of the other 65 tumors did (P < 0.001), indicating that MSI-driven inactivation of target genes modifies tumor chemosensitivity. Our observations allowed us to define the first useful predictive criteria for irinotecan response in patients with colorectal cancer.

The DNA mismatch repair (MMR) system is involved in the correction of base/base mismatches and insertion/deletion loops arising during replication. In addition, some of the MMR components participate in recombination and double-strand break repair as well as cell cycle regulation and apoptosis. The inactivation of MMR genes, usually hMSH2 or hMLH1, is associated with human colorectal cancers and is responsible for the characteristic microsatellite instability (MSI)+ phenotype of these tumors. Because MMR is assumed to modulate cytotoxicity to various chemotherapeutic agents that act upon DNA, our objectives have been to define its possible involvement in the cytotoxicity of topoisomerase inhibitors. We have shown that colorectal cancer cell lines defective in DNA MMR exhibit an increased sensitivity to both camptothecin, a topoisomerase I inhibitor, and etoposide, a topoisomerase II inhibitor. Sensitivity to these drugs cannot be predicted by measuring endogenous levels of topoisomerase I and II. Our results also indicate that neither p53 status, nor cell cycle alterations correlate with the sensitivity of colorectal cancer cells to topoisomerase inhibitors. On the other hand, our data showing that resistance to these drugs can be achieved by the functional complementation of hMLH1 in an hMLH1-defective cell line have allowed us to establish that MMR is a critical determinant for chemosensitivity. Interestingly, our observations provide the rationale for the better responsiveness of MSI+ tumors to CPT-11, a camptothecin derivative, which we have observed in patients with metastatic colorectal cancers.

AIMS:The aim of our study was to assess the potential relationships between tumor responsiveness to CPT11, an analogue of camptothecin, which selectively inhibits DNA topoisomerase I, and the microsatellite instability, a feature of tumors with DNA mismatch repair defect.METHODS:We designed a retrospective clinical study including 35 patients with metastatic colorectal cancer treated with CPT11, for which we analyzed the expression of hMLH1 and hMSH2 in the tumor and determined microsatellite status of repeated mononucleotide tracts present in the coding region of RII-TGFB, BAX, hMSH3 and hMSH6 genes.RESULTS:A partial or minor response was observed in 9 patients, disease stabilization in 14 patients and progression in 12 patients. Staining of hMLH1 was undetectable in 2 of the 35 tumors, while only 1 tumor lacked hMSH2 expression. Four of the 31 tumors analyzed displayed intragenic microsatellite instability. We found a good correlation between inactivation of TGFB-RII, BAX or hMSH3 genes and tumor response to CPT-11 (P =0.002).CONCLUSION:Our preliminary data suggest that intragenic microsatellite instability may influence tumor response to CPT-11 in patients with colorectal cancer.

Aims - The aim of our study was to assess the potential relationships between tumor responsiveness to CPT11, an analogue of camptothecin, which selectively inhibits DNA topoisomerase I, and the microsatellite instability, a Feature of tumors with DNA mismatch repair defect.Methods - We designed a retrospective clinical study including 35 patients with metastatic colorectal cancer treated with CPT11, for which we analyzed the expression of hMLH1 and hMSH2 in the tumor and determined microsatellite status of repeated mononucleotide tracts present in the coding region of RII-TGF beta, BAX, hMSH3 and hMSH6 genes.Results-A partial or minor response was observed in 9 patients, disease stabilization in 14 patients and progression in 12 patients. Staining of hMLH1 was undetectable in 2 of the 35 tumors, while only 1 tumor lacked hMSH2 expression. Four of the 31 tumors analyzed displayed intragenic microsatellite instability. We found a good correlation between inactivation of TGF beta -RII, BAX or hMSH3 genes and tumor response to CPT-11 (P = 0.002).Conclusion - Our preliminary data suggest that intragenic microsatellite instability may influence tumor response to CPT-11 in patients with colorectal cancer.