Letters to the Editor
To the Editor:
Alkushi et al. (1) are to be congratulated for their paper that identified significant differences in survival between patients with endometrial carcinoma who had a negative or weak positive reaction for p53 and patients with a strong positive p53 index. Using multivariate analysis, p53-expression was a significant prognostic indicator independent of age and FIGO stage. Significant differences in survival were demonstrated for patients with endometrioid carcinoma and patients with non-endometrioid carcinoma; p53-negative tumors had a favorable outcome whereas p53-positive tumors did not. The authors, however, did not present all the details of their data for p53 expression in endometrioid cancers with respect to tumor grade. We wonder if they could be persuaded to do so in the anticipation that their data may be of relevance in providing useful preoperative information. If their findings can be applied to women with grade 1 and 2, stage Ia or Ib endometrial carcinomas, patients with p53-negative tumors could be treated less aggressively than those with p53-positive tumors. This approach would be of particular relevance in the occasional premenopausal patient who wishes to maintain her fertility. At the other end of the spectrum, patients with advanced grade 3 tumors that were p53 positive might be spared unhelpful radical therapy and receive either novel therapy, as part of a controlled clinical trial, or palliative care.
A most exciting aspect of such a p53 test is its apparent robustness. Alkushi et al. carried out their study on microarrays, thus basing their data on a tiny sample of the tumor. Many diagnoses of endometrial cancer are now based on samples that represent as little as 4% of the cavity (2) and their results imply that the p53 reaction may still provide prognostic data from even a limited quantity of tissue. Additionally, because the authors rightly acknowledge the difficulties in determining appropriate thresholds to define prognostic subgroups using immunohistochemistry, they were able to demonstrate that most cases fell readily into subsets with either a p53 index of <25% or >75%. Their data suggests that the pathologist would only need to determine whether there is a reaction in a few cells or most of the cells. This would be a much easier task than, for example, the relatively complex assessments in breast carcinomas of staining intensity and estimates of the proportion of positive cells in the evaluation of estrogen and progesterone receptor assays and HER-2 expression (3).
Advances in cancer pathology mean that important therapeutic decisions may now hinge on detailed immunohistochemical analysis of prognostic markers performed in the routine diagnostic setting. If our hypothesis is correct, assessment of p53 status could be easily incorporated into routine diagnostic practice to distinguish women requiring more aggressive preoperative therapy. We hope that Alkushi et al will feel able to reveal more of their data and, to reverse a metaphor, open a Pandora’s box of new research hypotheses and therapeutic opportunity.
Mark K. Heatley
Pauline J. Carder
Department of Histopathology, St. James’ University Hospital, Leeds, England
1. Alkushi A, Lim P, Coldman A, et al. Interpretation of p53 immunoreactivity in endometrial carcinoma: establishing a clinically relevant cut-off level. Int J Gynecol Pathol
2. Rodriguez GC, Yaqub N, King M. A comparison of the pipelle device and the vabra aspirator as measured by endometrial denudation in hysterectomy specimens: The pipelle device samples significantly less of the endometrial surface than the vabra aspirator. Am J Obstet Gynecol
3. Ellis IO, Bartlett J, Dowsett M, et al. Updated recommendations for HER2 testing in the UK. J Clin Pathol