Background: Lung cancer causes approximately one million deaths each year worldwide and protein p53 has been shown to be involved in the intricate processes regulating response to radiation and/or chemotherapeutic treatment. Consequently, since antibodies against p53 (anti-p53 antibodies) are associated with mutations within the p53 gene it seems likely that these antibodies could, hypothetically, be correlated with prognosis.
Methods: Serum samples from patients with non-small cell lung cancer (NSCLC) admitted to the Department of Oncology, University Hospital, Uppsala, Sweden, during 1983-1996 were studied. Anti-p53 abs were measured using a sandwich ELISA (Dianova, Hamburg, Germany).
Results: The present study included 84 patients with stage IIIA-IV (advanced NSCLC). At least three serum samples from each patient were collected and altogether 529 serum samples were analysed for the presence of anti-p53 antibodies. The median value of anti-p53 antibodies was 0.06 (range 0 - 139.8). Seventeen percent of investigated NSCLC first serum samples (n = 84) expressed elevated levels of anti-p53 antibodies. Anti-p53 antibodies were not correlated to tumour volume or platelets. Survival analysis showed that anti-p53 antibodies were not associated with survival as revealed by univariate analysis (p = 0.29). However, patients with adenocarcinoma had a significantly poorer survival if they expressed anti-p53 antibodies (p = 0.01), whereas this was not found for patients with squamous cell carcinoma (p = 0.13). In patients where the blood samples were collected during radiation therapy, a statistically significant correlation towards poorer survival was found (p = 0.05) when elevated anti-p53 antibodies levels were present. No correlations to survival were found for serum samples collected prior to radiation therapy, during chemotherapy, or during follow-up. When anti-p53 antibodies were measured continuously, no increase in median anti-p53 values was observed the closer the individual patient come to death.
Conclusion: The result of the present retrospective study indicates that anti-p53 antibodies are not suitable for predictions concerning selection of patients with a more favourable outcome. Further prospective studies are, though, needed to fully elucidate this issue.