Disseminated cancer cells in the peritoneal cavity: what can we do when we detect them?

Abstract

washes obtained at surgery was recognized. Results of the examination are currently incorporated in the local stage classifi cation as the CY category, in which CY1 stands for positive cytology and classifi es the patient automatically into Stage IV [3]. The prognostic signifi -cance of free cancer cells in the peritoneal washes was also observed among patients who were entered onto the Dutch D1 vs D2 trial [4]. However, this was not suf-fi cient to arouse enough interest to cause Western inves-tigators to consider revising the Tumor Node Metastasis (TNM) classifi cation. One of the reasons may have been that accurate diagnosis of peritoneal washing samples through morphology of the fl oating cells was a formi-dable task. A more quantitative method may have been preferable but was not available at that time.Beginning in the early 1990s, several investigators turned to molecular diagnosis by immunostaining and reverse-transcriptase polymerase chain reaction (RT-PCR) to detect minimal residual disease and occult metastasis in various body components [5]. Of these, an attempt to apply the RT-PCR technique for detection of free cancer cells in the peritoneal cavity was a success [6] — much more so than detection in the bone marrow aspirates and peripheral blood whose association with clinical outcome was inconsistent for gastric cancer [7]. The relevance of detecting carcinoembyonic antigen (CEA) mRNA in the peritoneal washes as a prognostic determinant and predictor of peritoneal carcinomatosis was confi rmed not only through an independent valida-tion study by the same investigators [8] but more soundly among other institutions from numerous countries. The positive predictive value of CEA RT-PCR has not been fl awless [6], however, and the pursuit of an optimal target mRNA is ongoing. Several attempts have been reported using various target mRNAs, among which a combination of markers that are less sensitive but highly specifi c could be the most promising [9]. Ultimately, a good marker or set of markers may in due course be ling studies. Other issues that Cancer recurrence after surgery apparently is due to regrowth of microscopic residual disease that could neither be seen at surgery nor visualized with preopera-tive imaging studies. Gastric cancer is feared for its capacity to metastasize through various metastatic pathways, of which metastasis through the lymphatic pathway is the most frequently observed. However, it is believed in some countries, including Japan, that a certain degree of lymphatic metastasis can be controlled by conducting extended lymphadenectomy. Most sur-geons in Japan have actually seen numerous occasions when gastric cancer with pathologically confi rmed nodal metastasis is cured by D2 dissection alone, and are educated to believe that a failure to perform adequate nodal dissection will result in locoregional recurrence. Postoperative irradiation of the gastric bed would there-fore be a logical solution to augment a suboptimal surgery, of which D0 dissection is a notable example, and was actually found to signifi cantly improve survival of resectable gastric cancer in North America where locoregional recurrence is commonly observed after the curative surgery [1].Peritoneal metastasis, on the other hand, is a tough opponent. This could be a consequence of free cancer cells shed from the serosa of the T3 stage cancer. In addition, cancer cells could also be mechanically scraped off from the stomach or may leak out from the lymphat-ics during surgical manipulation. Peritoneal carcinoma-tosis has been reported to be the most frequent pattern of disease failure after D2 dissection [2], which, as men-tioned above, eradicates the cause for local recurrence. For several decades, Japanese surgeons have been exploring whether detection of free cancer cells in the peritoneal cavity could predict the risk for recurrence and hence lead to any alteration of the treatment strat-egy for that particular patient. Consequently, the rele-vance of a histopathologic examination of the peritoneal