Clinical prognostic factors of diffuse large B cell non-Hodgkin lymphoma: a retrospective study

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of NHL in Egypt. It represents about 49% of NHL presenting to the National Cancer Institute (NCI), Cairo University. CHOP regimen is the standard treatment used for NHL since the 1970s with only 30-40% overall survival. Recently, integration of Rituximab became a standard of care for patients with DLBCL. However, its widespread use in developing countries is still limited by the lack of financial coverage. Clinical prognostic factors, as well as the pathological markers, are mandatory to individualize treatment.

The aim of the study was to evaluate the clinical risk stratification models including the age adjusted International prognostic index (aaIPI), patients profile and dose intensity (DI) of Cyclophosphamide and Doxorubicin as effective tools for predicting the outcome and prognosis of our DLBCL patients treated with first line CHOP regimen.

This retrospective study included 224 patients with diffuse large B cell lymphoma who were treated with 3-8 cycles of CHOP regimen at the Medical Oncology Department, NCI, Cairo University during the time period from 1999 to 2006.

One hundred and seventy-eight patients (79.5%) achieved CR after the CHOP regimen with an observation period of 51 months. The median survival time was 12 months. The OS and DFS at 2 years were 82% and 68.8%, respectively. The univariate analysis of predictive factors for response to treatment showed that the CR rate was significantly affected by aa-IPI and its elements (performance status, stage & LDH), extranodal lesions and DI of Cyclophosphamide and Doxorubicin. The CR rate was 96.9%, 91.2%, 73.9% and 55.6% in cases with aa-IPI 0, 1, 2 and 3, respectively (p<0.001) and it was 82.4%, 81.9% versus 50% in cases with no extranodal site, one extranodal site and two extranodal sites, respectively (p=0.01). As regard DI of Cyclophosphamide, with DI below or equal to the median (249 mg/m(2)/week) the CR rate was 69%, while with DI above the median the CR rate was 87.7% (p=0.001). For Doxorubicin, the CR rate was 72.3% with DI below or equal to the median (16.5 mg/m(2)/week), however, it was 86.6% with DI above the median (p=0.008). The OS rate was significantly affected by aa-IPI as it was 89.8% in cases of aa-IPI 0+1 versus 75.8% in those of aa-IPI 2+3 (p=0.03). DI of Cyclophosphamide and Doxorubicin significantly influenced the OS. The OS rate was 74% with DI of Doxorubicin below or equal to the median versus 96% in cases with DI above the median (p=0.02). For Cyclophosphamide the OS rate was 72.7% with DI below or equal to the median versus 96.3% in cases with DI above the median (p=0.01). The tumor bulk (with a median tumor size of 5 cm) affected the OS, which was 91.23% versus 86.8% in the tumor bulk less than and more than or equal to the median, respectively (p=0.05). By multivariate analysis of predictive factors for response to treatment, the CR rate was significantly affected by the number of extranodal sites and the clinical staging of diffuse large B cell lymphoma. However, OS rate was strongly associated with the bulk of the tumor and the clinical staging of diffuse large B cell lymphoma.

DI of Cyclophosphamide and Doxorubicin is important in the future treatment regimen plan for DLBCL especially in high risk cases. In addition to aa-IPI and its elements, extra nodal sites and bulk of the tumor proved to be significant predictors and prognostic factors for DLBCL treatment outcome.