Prognostic impact of PCR-based identification of isolated tumour cells in the peritoneal lavage fluid of gastric cancer patients who underwent a curative R0 resection

Abstract

Identification of cancer cells in the peritoneal cavity could influence therapy and outcome of gastric carcinoma patients. The objective of this study was to evaluate the clinical impact of the real-time quantitative polymerase chain reaction-(PCR) based identification of isolated tumour cells in the peritoneal lavage fluid of gastric carcinoma. The peritoneal lavage fluid of 116 patients with gastric cancer was sampled at laparotomy. After RNA extraction and reverse transcription, real-time quantitative PCR was performed using the primers and probes for carcinoembryonic antigen (CEA) and cytokeratin-20 (CK20). When either the CEA mRNA or CK20 mRNA level of the sample was over the cutoff value, the sample was determined to be PCR-positive. Forty-six (40%) of the 116 patients were PCR-positive and 30 (65%) of the 46 PCR-positive patients died as a result of recurrent peritoneal dissemination. The prognosis of the 46 PCR-positive patients was significantly (Po0.001) worse than that of 70 PCR-negative patients. Furthermore, in 80 of the cases with a curative R0 resection, 15 of the patients with PCR-positive findings had a significantly (Po0.001) poorer prognosis than the 65 PCR-negative patients. The prognosis of the PCR-positive patients was significantly poorer than that of the PCR-negative patients in the T3 (Po0.0001) and T4 (P ¼ 0.048) subgroups. In a multivariate analysis of the 80 cases with a curative R0 resection, the real-time quantitative RT -PCR (CEA and/or CK20) levels indicated that they were independent prognostic factors. The real-time quantitative RT -PCR analysis of the CEA and/or CK20 transcripts in the peritoneal lavage fluid is useful for predicting the peritoneal recurrence in patients who are undergoing a curative resection for gastric cancer.

Carcinoembryonic antigen and cytokeratin 20 in peritoneal cells of cancer patients: are we aware of what we are detecting by mRNA examination?

“…In a recently published paper in your journal, Katsuragi et al (2007) show that peritoneal recurrence in patients with early-stage gastric cancer relates to the levels of carcinoembryonic antigen (CEA) and/or cytokeratin 20 (CK 20) transcripts in the peritoneal lavage specimens from these patients. The authors are convinced that a positive RT-PCR result identifies the presence of 'micrometastatic cells'.…”

“…The qRT-PCR-based techniques of CEA and CK 20 detection present the same limitations as conventional RT-PCR, as researchers were unable to set the cut-off values to distinguish between cancer and haematopoietic cell expression (Bustin et al, 2004;Schuster et al, 2004). Especially in the experimental setting studied by Katsuragi et al (2007), one should consider that CEA expression may be induced in peripheral blood mononuclear cells by cytokines (Jung et al, 1998;Goeminne et al, 1999), as increased concentrations of an array of cytokines characterise peritoneal fluids of cancer patients (Chechlinska et al, 2007). Furthermore, peritoneal lavage samples of cancer-free patients were found to contain CEA transcripts (Broll et al, 2001).…”

“…Unfortunately, none of these limitations, although widely discussed in numerous papers, have been considered or even discussed by Katsuragi et al (2007). In effect, the observed correlation between the positive CEA and CK 20 qRT-PCR signals and cancer recurrence seems reasonably well documented, whereas the statement that the authors detect 'free cancer cells' in their study, expressed already in the title and then subsequently used through the methods section to discussion, is totally unwarranted.…”

“…Considering the above, the measurements of the markers described by Katsuragi et al (2007) may reflect the presence of inflammatory cells rather than micrometastatic cells: the more so because as many as 35% of cytology-negative/PCRpositive patients developed no peritoneal metastases. Nevertheless, prognosis for PCR-positive patients was shown to be significantly worse than for PCR-negative patients, and this was in accordance with the recent data of Crumley et al (2006) and Deans et al (2007), who linked inflammation symptoms with adverse prognosis in patients with gastric cancer.…”

“…The study of Katsuragi et al (2007) is also methodologically flawed. The GAPDH gene was used as an internal control and a reference gene.…”