Immune parameters as predictors of response to maintenance therapy in low grade non-Hodgkin's lymphoma patients

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

The non-Hodgkin's lymphomas (NHL) are a heterogeneous group of malignancies characterised by the uncontrolled proliferation of lymphoid cells. The Working Formulation divides these neoplasms into low, intermediate and high-grade categories according to their natural history. Low grade NHL (LG-NHL) is clinically indolent whereas high grade NHL is more aggressive. Most LG-NHL patients respond well to chemotherapy, but are rarely cured, making death from LG-NHL virtually inevitable. In this study on LG-NHL patients, all patients received combination chemotherapy for 6 weeks, which consisted of cyclophosphamide, oncovin (vincristine) and prednisone (COP). After this treatment patients received either oral cyclophosphamide or alpha-interferon (α-IFN) as maintenance treatment for a period of two years. The ability to predict patients' treatment response at diagnosis would be extremely helpful for both the patient and the clinician. This would enable appropriate therapy to be instituted at diagnosis, whether it be cyclophosphamide or α-IFN, which might increase the subsequent survival of these patients, as this would prevent patients from receiving treatment to which they would not respond. Furthermore, patients who were unlikely to respond to either of these two treatments could be targeted for alternative treatment or form part of a clinical research trial. LG-NHL patients were assessed at diagnosis for a number of immune parameters. These included: - natural killer activity (NKA), α-IFN enhanced NKA, mitogen and antigen proliferation, lymphokine activated kill against both KS62 and Daudi targets, phenotypic analysis of circulating lymphocytes, assessment of IL-2 levels after mitogen stimulation of lymphocytes, as well as the determination of plasma IL-2 levels. All these above-mentioned parameters were serially monitored over time in an attempt to predict early relapse. A statistically significant reduction in percentage of circulating CD3+ and CD8+ cells and an increase in percentage NK cells was found in patients at diagnosis as compared to normal controls. A reduction in percentage of circulating NK cells over time appeared to be a good prognostic indicator and an increase in percentage NK cells a poor prognostic indicator in this group of patients, although this was not statistically significant. When the in vitro α-IFN enhanced NKA was indicated as a percentage of NKA, a negative correlation appeared to exist between this in vitro response to α-IFN and the in vivo response, although this was not statistically significant (i.e. those patients showing the least α-IFN enhanced NKA were those that responded clinically and those with the highest α-IFN augmented NKA either relapsed or transformed to a higher grade of NHL). By incorporating the percentages of circulating lymphocytes present in the peripheral blood, into the multivariate discriminant analysis, it was possible to derive formulae to enable the prediction of response to either α-IFN or cyclophosphamide. This was a particularly exciting and apparently novel finding. The work presented here has therefore established both a possible negative indicator of α-IFN response (a high in vitro α-IFN response) and a positive indicator (the formula generated in the multivariate discriminant analysis using the flow cytometric phenotypic analysis). By making using of both of these factors, it would increase the chances of patients being selected for appropriate forms of treatment and minimise the chances of patients suffering a relapse. The Kaplan-Meier curve indicated that cyclophosphamide treatment was more beneficial than α-IFN therapy in this group of patients, although this did not attain statistical significance. Verification of all of these above-mentioned findings would need further confirmation in a larger study.