1. Phase III Randomized Trial Assessing Rofecoxib in the Adjuvant Setting of Colorectal Cancer: Final Results of the VICTOR Trial Rachel S. Midgley, Christopher C. McConkey, Elaine C. Johnstone, Janet A. Dunn, Justine L. Smith, Simon A. Grumett, Patrick Julier, Claire Iveson, Yoko Yanagisawa, Bryan Warren, Michael J. Langman, and David J. Kerr J Clin Oncol 2010. vol. 28:4575-4580. R1. 박선희 / PROF. 이재진

2. INTRODUCTION Approximately half of all patients undergoing potentially curative surgery for colorectal cancer (CRC) ultimately experience relapse and die of metastatic disease.  introduction of adjuvant chemotherapy Cyclooxygenase-2 (COX-2) plays an important role in colorectal carcinogenesis during the transition from adenoma to carcinoma and in invasion, angiogenesis, and metastasis. –Immunohistochemical analysis of CRC tissue 70% of tumors express COX-2; ↑ with stage progression, coexpression of the progression factors matrix metalloproteinase-2 and vascular endothelial growth factor. –Haile et al : NSAIDs  reduction in polyp recurrence was greater in adenomas with high levels of COX-2 expression –Chan et al : aspirin use significantly reduced CRC risk in tumors over-expressing COX-2 but not in cancers with weak or absent COX-2 expression

3. Hypothesis : Rofecoxib would provide a safe approach to blocking COX-2 and reduce the rate of tumor recurrence in patients who had undergone potentially curative surgery for CRC and that the level of tissue expression of the COX-2 enzyme in the tumor might modulate this effect.

4. PATIENTS AND METHODS Patients –Inclusion criteria histologically proven stage II and III CRC in patients who had complete resection of the primary tumor WHO performance status of 0 or 1 hematologic, liver, and renal function within the normal range –Exclusion criteria Peptic ulceration or GI bleeding in the past year long-term nonsteroidal anti-inflammatory drug therapy younger than age 18 years history of cancer, inflammatory bowel disease, or severe congestive heart failure women who were pregnant or lactating premenopausal women not using contraception Trial Design –double-blind –Rofecoxib or placebo

5. Assessment of COX-2 Expression –Immunohistochemical staining –standardized grading system absent (score 0), weak staining (score 1), moderate staining (score 2), or strong staining (score 3) Statistical Considerations –Primary end point OS : random assignment ~ death from any cause. –Secondary end points DFS : random assignment ~ recurrence or death from any cause RFS : random assignment ~ recurrence and/or CRC death. –Kaplan-Meier curves, log-rank analysis, Cox proportional hazards models,

6. RESULTS Fig 1. Median time on Tx –rofecoxib : 7.4 mo. (3.1 ~ 14.0 mo.) –placebo : 8.2 mo. (3.7 ~ 14.9 mo.) Median f/u time –rofecoxib : 58.1 mo. (48.2 ~ 62.0 mo.) –placebo : 58.2 mo. (47.8 ~ 61.9 mo.) Early discontinuations in the rofecoxib arm  adverse effect : GI pain, heartburn, renal impairment, diarrhea, HTN, heart failure

9. OS HR for CRC-specific mortality : 0.98

10. DFS and RFS rofecoxibplacebo DFS event335369 3yr DFS75.6%73.4% recurrance297329 HR for DFS0.89 HR for RFS0.89

11. COX-2 Expression No treatment interaction (HR 0.75)

12. DISCUSSION Blockade of COX-2 by rofecoxib –direct inhibition of growth of residual micrometastases, prevention of angiogenesis, and reduction in tumoral expression of matrix metalloproteinases. Given the relative GI safety of rofecoxib, we initially reasoned that a small (3%) absolute improvement in 5-year survival, if proven, would be sufficient to warrant use of this agent in the adjuvant setting. Results show no overall benefit for rofecoxib, with no improvements in DFS or OS. N Engl J Med 2003;348:891-9.

13. why a statistically significant benefit was not established –drug is inactive –too few patients to demonstrate a clinically relevant but small effect –duration of effective COX-2 inhibition may have been insufficient to alter the malignant phenotype –inadequate representation of a COX-2–sensitive subpopulation. Data from the aspirin CRC prevention studies –a dose and duration effect for cancer risk reduction –positive adenoma prevention studies with COX-2 inhibitors, 16 ~ 36 months  The premature closure of this trial and abbreviated duration of treatment may have attenuated therapeutic benefit. From previous adjuvant studies –the greatest number of CRC recurrences occur in the first year after surgery  adjuvant chemotherapy benefits are accrued then, with little demonstrable therapeutic effect in subsequent years.  Significant trend toward prevention of recurrence by rofecoxib in the first year (110 v 141, P= 0.044)  more prolonged exposure to rofecoxib would be beneficial.

14. Tumoral expression of COX-2 is a poor prognostic feature Advanced NSCLC, increased COX-2 expression subgoup + celecoxib  therapeutic benefit J Clin Oncol 26:848-855, 2008 Prostate cancer, ↑COX-2 expression associated a poor prognostic feature Lancet Oncol 8:912-920, 2007 CRC, molecular background of liver meta  EGFR & COX-2 ↑ in metachronous & synchronous metastases Br J Cancer 99:1729-1734, 2008  This was not observed in this study Despite the RFS curves hinting at a similar effect, statistically there was no evidence of a treatment interaction for rofecoxib relative to COX-2 expression. There is still significant interest in the potential role of COX-2 inhibition in cancer secondary prevention, and a further trial of this family of drugs with careful patient selection, maintenance of exposure, and a careful parallel assessment of biomarkers is worthy of consideration.