Tumour Markers

destruction. These should be used in conjunction with estimations of conventional liver-related enzymes. In this way we may hope to combine as many independent variables as possible. The other twelve parameters gave disappointing results; in particular, calcitonin, the beta subunit of human chorionic gonadotrophin placental alkaline phosphatase and the polyamines were not found in abnormally high concentrations in more than a minority of patients with overt metastases. When the results in the patients without overt metastatic disease are examined, the situation is much less clear. Few of the markers are abnormal in more than one or two patients (Table 1) and of the seven most commonly elevated markers, many were normal at the time of mastectomy in the patients who have relapsed within eighteen months of sampling, although some were abnormal in those that relapsed. Because of the small number of patients it is not yet clear whether these parameters have any place in management, but it seems likely that by combining them in follow up of post-mastectomy patients they may have some predictive value. The other twelve parameters measured disclosed some abnormalities scattered throughout Groups 2 and 3, but appear to bear little relationship to prognosis. All the parameters studied here are relatively nonspecific and much more fundamental work will be needed to obtain a sensitive and specific tumour-index-substance for breast cancer.