Originally published as JCO Early Release 10.1200/JCO.2005.07.021 on September 26 2005

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hoff, P. M. Search for Related Content PubMed PubMed Citation Articles by Hoff, P. M.

Is There a Role for Routine p53 Testing in Colorectal Cancer?

Department of Gastrointestinal Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX

The availability of new and more effective chemotherapeutic and molecular targeted agents has resulted in improved treatment of regional and advanced colorectal cancer. However, the financial cost and the morbidity associated with the new treatments have increased as well, and to date no reliable marker capable of selecting patients who will benefit from a specific treatment has been identified. Individualizing treatment based on the molecular profile of the patient and the tumor could lead to considerable improvement in treatment cost and outcomes. Although all patients with colorectal cancer would benefit from better prognostic and predictive markers, risk-benefit is particularly important for candidates for adjuvant therapy, which may expose patients already cured by surgery to unnecessary risk or to a treatment that fails to maximize their chances of cure from micrometastasis.

The use of adjuvant oxaliplatin regimens after resection of stage II or III colon cancer is a good example of this clinical conundrum. The addition of oxaliplatin to fluorouracil and leucovorin improves the 3-year disease-free survival rate, and has been approved for high-risk resected colon cancer by the US Food and Drug Administration, but it is clear that not all patients benefit equally from oxaliplatin. In fact, when patients already cured by surgery alone and patients who experience disease recurrence despite treatment are excluded, fewer than 30% of stage II and III patients actually benefit from the use of any adjuvant chemotherapy. Unfortunately, we currently have no practical way of identifying those who will and those who will not benefit from the treatment. We still identify which patients should receive adjuvant chemotherapy by their pathologic staging, a tool that has changed little since originally proposed by Dukes more than 70 years ago.¹ The choice of the chemotherapy regimen to be used also lacks sophistication and the currently accepted regimens are applied to the entire patient population without consideration for individual tumor sensitivity or for the patient’s tolerance for the therapy.

Oncologists are coming to agree that this reality has to be changed. Toward this end, several molecular markers have been investigated, but none has been incorporated into routine treatment decisions. Early efforts to change this empiric approach to therapy are under way but have yet to produce results. For example, the Eastern Cooperative Oncology Group is finalizing the plans for E5202, a phase III trial in which the adjuvant treatment for high-risk stage II patients will be based on molecular markers, such as the presence or absence of 18q deletion and microsatellite instability.

The p53 gene is an important tumor suppressor gene located on the short arm of chromosome 17; somatic mutations of this gene are common in solid tumors. Its main task is to identify cells with significant DNA defects and coordinate cellular events that lead to cell cycle arrest in G₁ to allow time for repair or, if repair is not possible, apoptosis. Cells with mutant p53 proteins have reduced checkpoint activity and dysregulated cell proliferation. Given that avoiding apoptosis is crucial for cancer cells, p53 mutations are generally believed to have prognostic relevance. Loss of function of p53 is a late event in adenocarcinoma-carcinoma progression and has been identified in up to 60% of colon cancers.² Mutations can be detected either at the protein level by immunohistochemical analysis or at the gene level by sequencing; the latter technique is considered more accurate but is more laborious. The presence of p53 mutations is also thought to be predictive of decreased cancer cell sensitivity to most chemotherapeutic agents, particularly fluorouracil.³ However, results from available studies, which have been retrospective and generally underpowered, are conflicting with respect to prognostic and predictive roles in colorectal cancer.

In this issue of the Journal of Clinical Oncology, Russo et al⁴ present the results of a large retrospective analysis of p53 mutations in an attempt to confirm the role of p53 as a prognostic and predictive factor in colorectal cancer. This laudable effort used creative strategies to compile 3,583 patient cases from 25 research groups in 17 countries, thereby forming the largest pool of p53 mutation data in colorectal cancer ever published. Collaborating groups were identified by their published work using Medline and a specific Web page was created to facilitate interaction. The final result can be considered a truly global effort and this work highlights the collaborative research possibilities created by the astounding communication advances of recent years. An important feature of this analysis is that the authors examined gene mutation profiles rather than relying on protein expression as assessed by immunohistochemical analysis.⁵ In addition to being more reliable, sequencing of p53 allows for exploratory analysis of correlations between specific mutations and particular outcomes.

Despite the innovative design for data collection and the large sample size, there are some problems with this study. Unwanted biases and loss of statistical power are unavoidable in retrospective analyses, and data obtained from large samples should be viewed with caution, particularly when the results presented come from a pool of previously published work. In addition, the tumor samples were obtained from a heterogeneous group of patients. The diversity of the populations studied could have introduced other unanticipated molecular factors that influenced the results. Finally, the mutation analysis data were a compilation of the results provided by individual studies. Each contributing group conducted their own sequencing and identified p53 mutations by following different protocols with no unified standard method for specimen handling, preservation, preparation, and testing. Even relatively small differences in the handling and preparation of tumor specimens could have resulted in significant differences in the final results. These problems should not detract from the investigators’ efforts, but need to be taken into account when the results are being considered.

A major question for practicing oncologists is whether the results of this study support routine testing for p53 mutations. The short answer is no. Patients with colorectal cancer and p53 mutations had a nonsignificant trend toward worse survival that was more evident in distal tumors. This result is generally consistent with previous retrospective analyses.^5,6 However, the prognostic value was modest and not powerful enough to identify patients who should not receive adjuvant therapy. Testing for p53 mutations could be of value if it showed enough predictive power to help identify tumors more or less likely to respond to a specific antineoplastic treatment. In this review, the presence of p53 mutations failed to predict which patients would benefit from fluorouracil-based adjuvant chemotherapy. This result is not entirely surprising, given that a previous large retrospective study had yielded similar results.⁶

The study by Russo et al⁴ adds little to the available knowledge about the affect of p53 in patients with colorectal cancer, and although it generates some interesting hypotheses regarding the relationship between tumor sites and specific mutations that could be explored prospectively, it does not change current clinical practice. Ultimately, before a marker can be adopted as a meaningful prognostic or predictive tool, it needs to be standardized and validated prospectively, preferably as part of a randomized trial. More important, its use has to result in improved patient management and treatment selection. Although its routine use cannot be recommended, testing for p53 mutations remains of interest and research in the area should continue. Eventually, we may learn how to incorporate p53 as well as other prognostic and predictive markers into clinical practice. To provide the power necessary for clinical decision making, p53 mutation analysis probably will need to be incorporated in a larger panel of markers to result in a test with the desired prognostic or predictive power.

Except for the rare patients with Li-Fraumeni syndrome, who have germline mutations, most patients have somatic mutations of the p53 gene restricted to tumor cells. Therefore, designer treatments directed at killing only p53-mutated cells are being investigated. Other therapeutic areas of research include the development of gene therapies that could re-express wild-type p53. In summary, p53 is an important tumor suppressor gene that deserves additional investigation as a marker and as a therapeutic target in colorectal cancer, but which at present has no role in clinical practice.

Author's Disclosures of Potential Conflicts of Interest

The author indicated no potential conflicts of interest.

REFERENCES

1. Dukes C: The spread of cancer of the rectum. Br J Surg 12:643-648, 1930

2. Lane DP: Cancer: p53, guardian of the genome. Nature 358:15-16, 1992[CrossRef][Medline]

3. Bunz F, Hwang PM, Torrance C, et al: Disruption of p53 in human cancer cells alters the responses to therapeutic agents. J Clin Invest 104:263-269, 1999[Medline]

4. Russo A, Bazan V, Iacopetta B, et al: The TP53 Colorectal Cancer International Collaborative Study on the prognostic and predictive significance of p53 mutation: Influence of tumor site, type of mutation, and adjuvant treatment. J Clin Oncol 23:7518-7528, 2005[Abstract/Free Full Text]

5. Kressner U, Inganas M, Byding S, et al: Prognostic value of p53 genetic changes in colorectal cancer. J Clin Oncol 17:593-599, 1999[Abstract/Free Full Text]

6. Allegra CJ, Paik S, Colangelo LH, et al: Prognostic value of thymidylate synthase, Ki-67, and p53 in patients with Dukes' B and C colon cancer: A National Cancer Institute-National Surgical Adjuvant Breast and Bowel Project collaborative study. J Clin Oncol 21:241-250, 2003[Abstract/Free Full Text]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hoff, P. M. Search for Related Content PubMed PubMed Citation Articles by Hoff, P. M.