Background: Gastrointestinal cancer tumour markers are valuable in the detection of recurrence following resection or in monitoring response to chemotherapy. CEA, CA19-9, CA-50 and CA72-4 are currently available but are nonspecific and have a low sensitivity. ‘Tumour M2-pyruvate kinase’ was described by Eigenbrodt around 1985. In cancers the active tetrameric form of the M2 isoenzyme of pyruvate kinase converted to an inactive dimeric form by direct interaction with oncoproteins to channel glucose carbons into DNA synthesis. This review summarizes the current knowledge of this unique tumour marker with regard to its biochemistry, assay and potential use as a diagnostic and screening tool in gastrointestinal cancer.
Methods: A literature search was conducted for entries from 1980 to 2005 using PubMed and NeLH databases using tumour M2-pyruvate kinase, faecal tumour M2-pyruvate kinase, tumour metabolism, tumour markers and carcinoembryonic antigen as keywords. A total of 56 references relevant to tumour M2-pyruvate kinase were retrieved. Eighteen references were clinical studies involving plasma/faecal tumour M2-pyruvate kinase and gastrointestinal cancer. The remaining 38 references were clinical/nonclinical trials and reviews on tumour metabolism and plasma/faecal tumour M2-pyruvate kinase assay. Seven of the 18 clinical studies involved faecal M2-pyruvate kinase. Three of the 11 plasma tumour M2-pyruvate kinase studies were non-English language and were excluded. The sensitivity, specificity, positive predictive and negative predictive value for plasma/serum tumour M2-pyruvate kinase in the detection of gastrointestinal cancer was determined for each of the remaining eight studies. Data for gastrointestinal cancer M2-pyruvate kinase were compared with other gastrointestinal cancer markers. Data from three of the eight studies using a diagnostic cut-off value of 15 U/ml for ethylenediaminetetraacetic acid (EDTA) plasma tumour M2-pyruvate kinase were analysed together as a small meta-analysis.
Results: At a diagnostic cut-off value of 15 U/ml for tumour M2-pyruvate kinase in EDTA plasma the sensitivity, specificity, positive predictive and negative predictive value was 57.3, 89, 85.7 and 64.8%, respectively, for colorectal cancers, 62.1, 89, 88 and 64%, respectively, for gastric/oesophageal cancers and 72.5, 89, 58 and 94%, respectively, for pancreatic cancers. As a faecal marker for colorectal cancers, faecal tumour M2-pyruvate kinase has a sensitivity of 73–92% at a cut-off value of 4 U/ml as against 50% sensitivity for Guaiac faecal test.
Conclusion: Circulating tumour M2-pyruvate kinase is more commonly elevated in oesophageal, gastric and colorectal cancer patients than conventional tumour markers. Faecal M2-pyruvate kinase is a sensitive marker of colorectal cancer. The clinical role of tumour M2-pyruvate kinase in gastrointestinal cancer management should be investigated in large-scale clinical trials.
Department of Surgery, Royal Free Hospital, Royal Free and University College Medical School, London
Correspondence to Professor Brian R. Davidson, FRCS, MD, University Department of Surgery, 9th Floor, Royal Free and University College Medical School of UCL, Pond Street, London NW3 2QG, UK
Tel: +44 20 85273081; e-mail: firstname.lastname@example.org
Received 3 March 2006 Accepted 25 August 2006