May 1, 2006 - Abstract Purpose: Colon cancer cells with high-frequency microsatellite instability (MSI-H) display resis- tance to 5-fluorouracil (5-FU) that can ...
Human Cancer Biology
Thymidylate Synthase Expression in Colon Carcinomas with Microsatellite Instability Frank A. Sinicrope, Rafaela L. Rego, Kevin C. Halling, Nathan R. Foster, Daniel J. Sargent, Betsy La Plant, Amy J. French, Carmen J. Allegra, John A. Laurie, Richard M. Goldberg, Thomas E.Witzig, and Stephen N. Thibodeau
Abstract
Purpose: Colon cancer cells with high-frequency microsatellite instability (MSI-H) display resistance to 5-fluorouracil (5-FU) that can be reversed by restoring DNA mismatch repair (MMR) proficiency. Given that thymidylate synthase (TS) is inhibited by 5-FU, we studied the relationship between MSI and TS expression, and the prognostic effect of these and other markers (i.e., p53 and 17p allelic imbalance). Experimental Design: Dukes’ stage B2 and C colon carcinomas (n = 320) from participants in 5-FU-based adjuvant therapy trials were analyzed for MSI and17p allelic imbalance. Expression of MMR (hMLH1, hMSH2), TS, and p53 proteins were analyzed by immunohistochemistry. Correlations between markers and associations with overall survival were determined. Results: Of 320 cancers studied, 60 (19%) were MSI-H. TS expression variables were similar in MSI-H and microsatellite stable/low-frequency MSI (MSS/MSI-L) cancers, and unrelated to MMR proteins. MSI-H tumors had lower stage (P = 0.0007), fewer metastatic lymph nodes (P = 0.004), and improved overall survival (P = 0.01). Loss of MMR proteins was also associated with better overall survival (P = 0.006). None of theTS variables were prognostic. Histologic grade (P = 0.0008) and nodal status (P = 0.0002) were associated with overall survival, in contrast to 17p allelic imbalance or p53. Only MSI status or loss of MMR proteins, histologic grade, and tumor stage were independent markers for overall survival. Conclusions: MSI-H tumors show earlier stage at presentation and better stage-adjusted survival rates. MSI status and TS expression were unrelated and TS was not prognostic, suggesting thatTS levels cannot explain therapeutic resistance to 5-FU reported in MSI-H colon cancers.
Colorectal cancer is the fourth most incident cancer in the United States and is the second leading cause of cancer-related mortality (1). The majority of sporadic colorectal cancers show chromosomal instability due to loss of heterozygosity and DNA aneuploidy (2, 3). Another pathway that accounts for up to 20% of colon cancers is characterized by microsatellite instability (MSI). MSI is caused by defective DNA mismatch repair (MMR), and in sporadic cases is nearly always due to hypermethylation of the hMLH1 gene promoter (4). Colon cancers with high-frequency MSI (MSI-H) are characterized by proximal location, frequent DNA diploidy, and pathologic features that commonly include poor differentiation, mucinous histolAuthors’ Affiliation: Mayo Clinic and Mayo College of Medicine, Rochester, Minnesota Received 1/24/06; revised 2/17/06; accepted 2/20/06. Grant support: National Cancer Institute grant CA104683 (F.A. Sinicrope). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Requests for reprints: Frank A. Sinicrope, Gastroenterology/Hepatology, Mayo Clinic, 200 First Street Southwest, Rochester, MN 55905. Phone: 507-286-8660; @ mayo.edu. Fax: 507-284-9111; E-mail: sinicrope.frank@ F 2006 American Association for Cancer Research. doi:10.1158/1078-0432.CCR-06-0178
Clin Cancer Res 2006;12(9) May 1, 2006
ogy, and peritumoral lymphocytic infiltration (5 – 7). In colon cancer cell lines with defective DNA MMR, resistance to the cytotoxic effects of 5-FU have been observed in vitro (8). Specifically, the HCT-116 colon cancer cell line contains a mutated hMLH1 gene and displays resistance to 5-FU. Transfer of chromosome 3 containing the intact hMLH1 gene into these cells was shown to restore 5-FU sensitivity (9). Furthermore, a 5-FU-resistant cell line displaying MSI due to hypermethylation of the hMLH1 promoter was rendered sensitive after demethylation and reexpression of hMLH1 using 5-azadeoxycytidine (10). The potential clinical relevance of these findings for MSI-H were shown by Ribic et al. (11), wherein patients with resected Dukes’ stage B2 and C colon cancers with MSI-H did not benefit from 5-FU-based adjuvant therapy, whereas microsatellite stable/low-frequency MSI (MSS/MSI-L) patients showed a survival advantage. These retrospective data, although not definitive, suggest that molecular features associated with the MSI-H phenotype may confer resistance to 5-FU. 5-Fluorouracil (5-FU)-based treatment remains the standard of care for the adjuvant treatment of resected colorectal cancer where it is combined with oxaliplatin (12 – 14) and for metastatic disease where it is coadministered with either oxaliplatin or irinotecan (15). Fluoropyrimidines irreversibly inhibit the thymidylate synthase (TS) enzyme, leading to DNA damage
2738
www.aacrjournals.org
Downloaded from clincancerres.aacrjournals.org on October 21, 2015. © 2006 American Association for Cancer Research.
Thymidylate Synthase Expression MSI-H in Colon Cancers
and blockade of DNA replication and repair (16). TS catalyzes conversion of dUMP to dTMP, the latter being necessary for DNA synthesis. Overexpression of TS protein or mRNA levels in primary colorectal cancers, as determined by immunohistochemistry or reverse transcription-PCR, have been associated with reduced responsiveness to 5-FU and adverse outcome in many (17 – 21), but not all, studies in patients with metastatic disease (22, 23). In the adjuvant setting, the relationship between TS protein levels and patient survival rates after 5-FU-based treatment is less well established (21, 24 – 26). Differences in the molecular profile of MSI-H versus MSS/MSI-L colorectal cancers may contribute to potential differences in their responsiveness to the cytotoxic effects of 5-FU. However, whether such differences include TS expression is largely unknown and remains a highly relevant clinical issue. p53 has been shown to transcriptionally regulate TS (27) and alterations in TS levels may contribute to the effect of p53 functional status upon 5-FU sensitivity. p53 is a known negative regulator of the cell cycle via its downstream effector protein p21WAF1/CIP1, a cyclin-dependent kinase (28). p21WAF1/CIP1 inhibits the cell cycle after DNA damage (29) and was found to be overexpressed in a higher proportion of MSI-H compared with MSS/MSI-L colorectal cancers (30 – 32). Metastatic colorectal cancers with wild-type p53 were found to have significantly lower TS levels compared with tumors with mutated p53 (18). Furthermore, MSI-H colorectal cancers have reduced rates of p53 mutation (33 – 35) and nuclear p53 protein expression compared with MSS/MSI-L tumors (36, 37). The potential relevance of these findings are underscored by the observation that wild-type p53 was associated with an improved response and prolonged survival in colon cancers after 5-FU-based adjuvant therapy (11, 38, 39), as were tumors that retained heterozygosity at either 17p or 18q (40). In this study, we analyzed the relationship between the level of TS protein expression and MSI status or MMR proteins with clinical outcome in Dukes’ B2 and C colon cancer patients treated in 5-FU-based adjuvant therapy trials. We also examined TS levels in relation to p53 expression and chromosome 17p allelic imbalance.
Materials and Methods Primary colon carcinomas were analyzed from patients who underwent surgical resection and who participated in one of five 5-FU-based adjuvant therapy trials conducted by the North Central Cancer Treatment Group. Paraffin-embedded tumor tissue was available from a nonrandom subset of study participants and included Dukes’ stage B2 (n = 54) and C (n = 266) colon cancers (total n = 320). Tumor histologic grade was defined by the American Joint Committee on Cancer Prognostic Factors Consensus Conference (41) as follows: grade 1, well differentiated; grade 2, moderately differentiated; grade 3, poorly differentiated; grade 4, undifferentiated. All patients were censored at 5 years after randomization for disease-free survival. Patients were followed for a minimum of 8 years after study randomization for overall survival data. Details of the five completed, randomized 5-FU-based adjuvant chemotherapy trials have been previously reported (25). The current analysis was in accordance with the original informed consent document. Of 320 patients, 233 were randomized to treatment arms and 87 received observation alone. Given that 21 patients received treatment that was later determined to be ineffective (i.e., portal venous 5-FU), there were a total of 212 patients who received effective
www.aacrjournals.org
treatment that included 5-FU. Accordingly and for purposes of the analyses, treatment was categorized as none or ineffective [n = 108; observation only (n = 87) + portal venous 5-FU (n = 21)] versus effective (i.e., consisting of modulated 5-FU; n = 212). MSI testing and 17p allelic imbalance. Tumors (n = 320) had been analyzed for MSI and allelic imbalance using 11 dinucleotide microsatellite markers as previously described (7, 42). Chromosome 17p allelic loss was determined using markers that included D17S261 and TP53 on 17p at or near the p53 locus. The median number of markers that gave a PCR product for both normal DNA and tumor DNA was nine (range, 5-11). Tumors were classified into three groups: (a) MSS with no MSI at any of the loci examined; (b) low instability (MSI-L, 3 Unknownc Dukes’ stage B2 C Treatmentx None or ineffective Effective MMR proteinsk Absent Intact
No. patients
5-y DFS (%)
P*
5-y OS (%)
P*
77 (24%) 236 (74%) 7 (2%)
74.0 62.3
0.2125
74.0 64.4
0.9987
123 (38%) 177 (56%) 13 (4%) 7 (2%)
62.6 66.1
0.7787
64.2 67.8
0.7354
125 (39%) 188 (59%) 7 (2%)
67.2 63.8
0.6542
68.0 66.0
0.4963
135 (42%) 184 (57%) 1 (1%)
69.6 63.0
0.1919
71.1 64.7
0.6891
36 (11%) 80 (25%) 204 (64%)
72.2 55.0
0.1341
72.2 55.0
0.9113
164 (51%) 129 (40%) 27 (9%)
60.4 74.4
0.0072
62.2 76.0
0.0365
260 (81%) 60 (19%)
63.8 75.0
0.0714
65.0 78.3
0.0115
221 (69%) 99 (31%)
71.0 54.5
3 Dukes’ stage B2 C
MSS/MSI-L
MSI-H
P*
59 (23%) 196 (77%)
18 (31%) 40 (69%)
0.2075
101 (41%) 144 (59%)
22 (40%) 33 (60%)
0.8675
100 (39%) 155 (61%)
25 (43%) 33 (57%)
0.5853
114 (44%) 146 (56%)
42 (70%) 18 (30%)
0.0003
94 (36%) 165 (64%)
41 (68%) 19 (32%)
