Date of Award

January 2013

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Michael Girardi

Subject Area(s)



While the spectrum of clinical tools used for blood assessment in cutaneous T cell lymphoma (CTCL) is vast, there is substantial variability in these practices between institutions. Furthermore, few studies have compared the markers for blood involvement for measures of performance (sensitivity, specificity, positive predictive value, negative predictive value). With the limited guidelines currently in place regarding blood testing protocols, we hypothesized that clinical practices between institutions is characterized by considerable heterogeneity. Given the prognostic importance of blood involvement in CTCL, we endeavored to 1) determine the spectrum of clinical practices with regards to assessment of blood in CTCL and 2) determine, which, if any of the markers most frequently used to identify abnormalities in the peripheral blood possess superior performance measures. A survey study was initiated to members of the USCLC to characterize clinical practices for analyzing blood in CTCL. Concurrently, a retrospective analysis of peripheral blood analyses from 150 patients seen at the Yale Cutaneous Lymphoma Center was conducted to compare the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for individual parameters (CD4/CD8 ratio ≥10, %CD4+/CD26- ≥ 30%, %CD4+/CD7- ≥ 40%, TCR gene rearrangement via PCR, TCR Vβ restriction) as compared with a gold standard of ISCL-criteria for B2 blood rating. Thirty-three institutions responded to the survey. Fifteen percent of responders indicated utilizing Vβ analysis, 80% utilized PCR, 100% utilized flow cytometry, and 55% performed Sézary cell enumeration. Screening for blood involvement began as early as stage IB for over 50% of respondents. Individually, the parameter CD4+/CD7- ≥ 40% possessed the highest specificity and PPV for blood involvement while %CD4+/CD26- ≥ 30% possessed the highest sensitivity and NPV. However, within the population of patients with a detectable Vβ restricted clone, the specificity and PPV of all three parameters (CD4/CD8 ratio ≥10, %CD4+/CD26- ≥ 30%, %CD4+/CD7- ≥ 40%) was 1.0. Hence, it appears that there is currently significant inter-institutional variation in clinical practices used to detect blood involvement in CTCL, which highlights the importance of continuous and judicious evaluation of the reliability and performance measures of the tools, old and new being used. As Vβ appears to improve the specificity and PPV of typical SS parameters and demonstrated good concordance with immunophenotypic and molecular techniques, it may be a valuable addition to the panel of studies obtained when considering CTCL patients for blood involvement.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.