Validation of international prognostic index for non-hodgkin's lymphoma in northeast Peninsular Malaysian Malays

Loading...
Thumbnail Image
Date
2002
Authors
Abdullah, Abu Dzarr Ganesh
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
International prognostic index (IPI) was introduced to risk stratify non-Hodgkin lymphoma (NHL) patients and to identify high-risk patient who might not respond favorably to standard chemotherapy. IPI was modeled from a Caucasian based patient population. We undertook a single center, observational longitudinal study involving all available patients with aggressive NHL who had received treatment from Hospital USM between 1st Jan 1990 and 31st Dec 2000. We confined our study to adult Malay patients to test the applicability of IPI in this racial group. Individual presenting clinical features was categorized as in the IPI study, and the patients' outcome in terms of complete response (CR), overall survival (OS) and disease free survival (DFS) rates for each of the above features were determined. At the same time the available data was used to characterize NHL disease profile in our patient population. From a total of 102 patients with NHL, only 88 patients had readable histopathology, in which 75 (85o/o) was diffuse aggressive NHL. T-celllymphomas constitute 29o/o of all NHL with immunophenotype studies. A final cohort of 45 Malay patients with aggressive NHL was established, 51% were of primary extra-nodal lymphoma. W aldeyer' s ring was the commonest extra-nodal site involved (18%) followed by gastrointestinal tract, bone and marrow llo/o each. 51% and 33% achieved complete clinical response (CR) and partial response (PR) correspondingly following CHOP chemotherapy. Overall response rate was 84%. 3-years survival rate was 45% and 5-years survival rate was 24%. Individual clinical characteristics were studied in terms of its association with complete response rate (CR), overall survival rates (OS) and disease free survival rates (DFS). Only good performance status (ECOG 0 & 1) were found to be significantly associated with better CR rates compared to poor perfonnance status (ECOG 2 to 4) p=0.007. RR=3.0 (95%CI 1.27 - 7.14). Male sex, age less than 60, limited disease, non bulky disease, good performance status, low hemoglobin, normal serum lactate dehydrogenase (LDH) and albumin were associated with higher CR rates but statistically not significant. Good performance status, presence of B symptoms and elevated serum LDH (>500iu/L) were shown to be significantly associated with longer OS (P= 0.0001, 0.02, 0.037) compared to poor performance status, absence of B symptoms and normal serum LDH. Early achievement of clinical complete response following 3rd cycle of chemotherapy was also associated with longer overall survival (P=0.017). With Cox regression analysis, only good performance status was significantly associated with overall survival. (P=0.01). The hazard ratio for good performance status was 0.242 (95o/oCI: 0.083- 0.71). We were not able to show statistically significant association between anemia with CR, OS and DFS rates due to poor statistical power. Only female gender was shown to have a significant association with bt(tter DFS rates (P=0.034) but this outcome was erroneous due to small patient numbers, il_l which survival data from the female gender group only consisted of censured data. In summmy, due to limited number of availaple patient for this study, we lacked statistical power to test our hypothesis. Only performance status was proven to prognosticate survival risk. Our study showed similar trends in survival outcome for other factors proposed in IPI, mimicking the trends seen in the IPI model. However these trends were not marked enough to become statistically significant in a poorly powered study. At this juncture, we conclude that there is no evidence to show that IPI is not applicable in our patient population. However, in the presence ofhigh proportion of primary extra-nodal lymphoma the modified three tiers IPI suggested for Chinese population may be a more appropriate option compared to the four tiers IPI.
Description
Keywords
International prognostic index (IPI)
Citation