Prognostic Significance of Quantitative Viral Markers in Adult T-Cell Leukemia/Lymphoma

To the Editor:

Adult T-cell leukemia/lymphoma (ATL) is a T-cell non-Hodgkin’s lymphoma frequently associated with a leukemic phase and is caused by human T-cell lymphotropic virus type I (HTLV-I).1Development of ATL is preceded by high HTLV-I antibody titers2 and characterized by monoclonal integration of proviral DNA in mononuclear cells in peripheral blood and/or lymph nodes.3 ATL has a broad clinical spectrum with several subtypes and corresponding differences in survival.4 The acute and lymphoma subtype have a prognosis of less than 1 year, while chronic and smoldering have longer survival and frequently precede the more severe types. Because proviral DNA level (proviral load) and antibody titers are important diagnostic markers, we evaluated their utility as predictors of survival in ATL.

We analyzed 30 ATL patients (acute, lymphoma, chronic subtype) from the Non-Hodgkin’s Lymphoma Registry at the University of the West Indies in Jamaica. All participants were enrolled in studies approved by Protocol and Human Subjects’ Review Committees at the National Cancer Institute (Bethesda, MD) and the University of the West Indies (Kingston, Jamaica).

Serum samples were screened by whole-virus enzyme-linked immunoassay (EIA) (Dupont, Wilmington, DE) and confirmed by Western blot (Biotech, Rockville, MD). HTLV-I antibody titers were assayed by the end-point dilution method using an EIA (Genetic Systems, Seattle, WA, or Cambridge-Biotech, Rockville, MD) at fourfold dilutions. Quantitative proviral DNA levels were detected by a real-time automated polymerase chain reaction (PCR) method. Ten microliters of DNA was amplified for 45 cycles with AmpliTaq Gold polymerase using an ABI PRISM Sequence Detection System and TaqMan PCR Reagent [P/N N808-0230] (PE Applied Biosystems, Foster City, CA) in a 96-well format. The HTLV-I/II primers were from highly conserved sequences (GenBank National Center for Biotechnology Information, Bethesda, MD) from the tax gene [HTV-F5 (7358-7378) and HTV-R4 (7518-7499)]. Triplicate reactions were performed and unknown copy numbers were automatically calculated by interpolation from a plasmid control regression curve and reported as copy equivalents per 10⁵ lymphocytes. The assay reliably detects at least 3 copies per 10⁵ lymphocytes. Samples with undetectable virus were scored as 1 copy per 10⁵ lymphocytes for calculations.

Kaplan-Meier life-table methods were used to estimate survival and 95% confidence intervals. The log-rank statistic was used to evaluate differences between survival curves. A dichotomized variable above or below the median value was used to compare survival by white blood cell count, proviral DNA, and antibody titer levels at diagnosis. Lymphocytosis was defined as lymphocyte count exceeding 4,000/μL. Each numeric value was log₁₀ transformed and ultimately reported as arithmetic values. Kruskal-Wallis test was used to compare mean values between groups. Correlations were examined using Spearman’s rank order statistic. All P values were two-sided.

Median survival was significantly shorter for acute (101 days, 95% confidence interval [CI] [54 to 124 days]) and lymphoma (83 days, 95% CI [63 to 314 days]) subtypes compared with the chronic subtype (P = .009) whose median survival could not be determined because 3 of 6 patients were still alive. Significant adverse clinical prognostic factors at diagnosis included hypercalcemia, hepatomegaly, splenomegaly, and B symptoms. Other factors that were not significant included white blood cell count and lymphocytosis. Proviral load and antibody titers were also not significant. Differences in proviral load between subtypes were significant (P = .01), with higher levels among acute and chronic subtypes compared with lymphoma subtype (Table1). Interestingly, acute ATL patients had a higher white blood cell count compared with the chronic and lymphoma subtypes. The elevated white blood count among acute ATL cases did not correspond to the highest proviral load, although white blood cell count was significantly correlated with proviral DNA level in all patients (R = .47; P = .009). Chronic and acute ATL patients had similar median absolute lymphocyte counts (9,679 and 8,742 cells/μL, respectively) and about 50% of both subtypes had circulating abnormal lymphocytes. These data suggest that chronic ATL patients may have more detectable virus per cell. Antibody titer levels were similar between the subtypes (P = .23). However, a trend toward better survival, 471 days (95% CI 102 to 935) versus 83 days (95% CI 54 to 314), was observed for cases with titer levels above the median (1:29,765) at diagnosis.

Viral markers were not significant predictors of survival among ATL patients, although others have shown that the ability to detect proviral DNA declines in complete remission after therapy.3In this study, higher antibody titers at diagnosis appeared to be associated with longer survival. Chronic ATL patients with high antibody titer levels, reflective of a heightened immune response, either require no immediate therapy or respond well to various treatment approaches,5 despite highest proviral load. The lymphoma subtype characterized by a low level of circulating malignant cells had the lowest proviral load but poorest survival. This may explain why these cases would fail to respond to anti–viral-containing regimens such as zidovudine and interferon-α6,7 and other immunotherapeutic approaches.5 To improve response rates among ATL patients, targeted, tumor-specific therapy that also enhances immunity is needed.