Analysis of GD2/GM2 synthase mRNA as a biomarker for small cell lung cancer.

GD2/GM2 synthase is a key enzyme in the synthesis of GD2 and GM2 gangliosides found on the surface of neuroblastoma and small cell lung carcinoma (SCLC) cells. In neuroblastoma, persistent levels of GD2/GM2 synthase RNA in bone marrow (BM) following therapy portend poorer progression-free and overall survival. We conducted this study to determine if GD2/GM2 synthase RNA could be detected in SCLC cell lines and human tissues, and whether mRNA transcript levels corresponded with disease status. Initially, a pilot study enrolled patients with SCLC to determine the rate of GD2 expression at various points in the patients' disease course. Peripheral blood (PB), bone marrow and tumor tissues were used to measure GD2/GM2 synthase levels. In addition, SCLC cell lines were analyzed for GD2/GM2 synthase expression. Based on data from that initial analysis, a prospective trial was developed enrolling patients with newly diagnosed SCLC and following them serially. GD2/GM2 synthase transcript was determined by a sensitive quantitative reverse transcription-PCR (qRT-PCR) assay and normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Six SCLC cell lines were assayed for expression of GD2/GM2 synthase, and high expression was detected in all. GD2/GM2 synthase transcript levels were obtained from tumor tissue, BM, or PB of 29 patients in the pilot study. 6/10 (60%) tumor tissues or BM samples were positive (median 332.7 units; range 13-2323 units); 8/19 (42%) untreated patients were GD2/GM2 synthase positive in their PB prior to beginning therapy (median 10.2; range 5.1-32.2); 3/4 (75%) patients who were first tested when they developed recurrent disease were positive in their PB (median 16.1; range 8.5-19.9). The fourth patient had an initial value of 2.0 (negative), which increased to 8.4 (positive) within 1 month without treatment. Seven of 12 patients with baseline positive GD2/GM2 synthase values had post-treatment levels measured, all of which were </=5 or showed a >50% decrease following successful treatment. Patients in the prospective trial demonstrated lower rates of positivity, with only 3/26 (12%) patients exhibiting detectable transcript levels in the peripheral blood prior to treatment. All 3 of these patients had their transcript levels fall below 5 after treatment. 11/26 patients had baseline levels of zero. Bone marrow was drawn at baseline on 7 patients in the prospective trial and 3 (43%) had transcript levels above 5 (range 0.65-27.43 units). There was no correlation between elevated levels in the BM and elevated levels in the PB. Although initial studies demonstrated that GD2/GM2 synthase transcripts were measurable in the peripheral blood of SCLC patients at diagnosis and declined with successful treatment, in a separate prospective study, these results could not be confirmed. Thus, GD2/GM2 is not a reliable biomarker in SCLC.