Analysis of T-cell receptor gene rearrangement for predicting clinical outcome in patients with cutaneous T-cell lymphoma: a comparison of Southern blot and polymerase chain reaction methods

Abstract

Objective: To extend previous observations regarding the prognostic value of analyzing lymph node DNA from patients with cutaneous T-cell lymphoma for the presence of a monoclonal T-cell population by Southern blot vs polymerase chain reaction (PCR) methods.

Design: Inception cohort study from 1982 to 1998. Recruitment of new patients ended in 1994.

Setting: A tertiary care referral center in Seattle, Wash. Patients Fifty-five uniformly staged patients with the diagnosis of cutaneous T-cell lymphoma who underwent a lymph node biopsy, 21 with clinically abnormal nodes and 34 with normal nodes. Interventions Lymph nodes were evaluated for T-cell receptor (TCR) gamma-chain gene rearrangement by 2 PCR methods: capillary electrophoresis and denaturing gradient gel electrophoresis. The same lymph nodes were evaluated by Southern blot analysis for TCR beta-chain gene rearrangement and examined histopathologically on the basis of the National Cancer Institute lymph node classification system. Patients were observed clinically for a mean of 9.5 years.

Main outcome measures: Skin stage, clinical lymph node examination, lymph node histologic examination, Southern blot analysis, and PCR analyses were evaluated as potential prognostic predictors by univariate and multivariate analyses. The statistical association of TCR analysis and clinical outcome was determined among all patients. Hazard ratios (HRs) by Cox proportional hazards regression analysis were used to estimate the risk of a poor clinical outcome. Cumulative survival rates were analyzed by the Kaplan-Meier method.

Results: A skin stage of T3 (tumors) or T4 (erythroderma) was the most powerful predictor of a poor clinical outcome (HR, 31.3 vs T1; P<.001). Patients with detectable TCR gamma-chain gene rearrangement in lymph node DNA by PCR also were more likely to have a poor outcome (HR, 5.1; P<.001), but it was a less powerful predictor than skin stage. Even when the skin stage, presence or absence of lymphadenopathy, and histologic lymph node score were known for the patient, Southern blot analysis still added to prediction of a poor outcome (HR, 9.3; P = .007), whereas PCR provided no statistically significant additional information on outcome.

Conclusions: Detection of a monoclonal T-cell population by PCR in lymph nodes of patients with cutaneous T-cell lymphoma does not enhance prediction of clinical outcome and probability of survival beyond what can be determined from clinical examination and histologic lymph node scores. Skin stage and the presence or absence of lymphadenopathy remain the most important determinants of clinical outcome.