Rituximab Fails to Reduce Histologic Transformation (HT) Rate of Follicular Lymphoma (FL) to Diffuse Large B-Cell Lymphoma (DLBCL)

Background: Histologic Transformation (HT) of indolent follicular lymphoma to aggressive lymphoma is a critical and sometimes fatal event in a patient’s disease course. It is unclear if rituximab influences HT. We have previously shown that single agent rituximab is used earlier in the disease course than traditional chemotherapy likely due to rituximab’s high therapeutic index (Cohler et al., ASH 2007). Others have shown rituximab also improves the complete response rate, disease free and overall survival of follicular lymphoma when used as a single agent and in combination with chemotherapy. Theoretically, this could reduce the disease burden over a patient’s lifetime and consequently the transformation rate.

Methods: We retrospectively screened 3500 patients and identified 584 eligible patients at Memorial Sloan Kettering Cancer Center (MSKCC) with newly diagnosed, treatment naïve, indolent follicular lymphoma. We compared two cohorts of rituximab usage: 1998– 2000 and 2001–2006. In the former, patients received rituximab predominantly in the relapsed setting. In the latter, patients liberally received rituximab even as single agent first line therapy. Histologic transformation to diffuse large B cell lymphoma (DLBCL) was the primary study endpoint. All therapy was recorded.

Results: Median follow-up time was 92 months for the 1998–2000 cohort and 41 months for the 2001–2006 cohort. Patients in the latter cohort received rituximab both earlier in the course of follow up and more often as a first line therapy. The median time from diagnosis to first rituximab was 21 months vs. 2 months, respectively. Rituximab was given alone or in combination as first line systemic therapy to 36% of the 1998–2000 cohort, but to 93% of the 2001–2006 cohort. The comparative risks of transformation between the two cohorts were not statistically significant (P-value = 0.41 by log rank comparison). The cumulative incidence of transformation 36 months after diagnosis was 8.1% for the 1998– 2000 cohort and 4.4% for the 2001–2006 cohort. Furthermore, patients receiving rituximab first line, either single agent or in combination, compared to patients receiving rituximab as salvage therapy, showed essentially no difference in risk of histologic transformation. (P-value = 0.68) Surprisingly, patients never receiving rituximab had a significantly lower risk of transformation than those who received rituximab at any point (p-value = 0.0095), however, these rituximab naïve patients had lower risk FLIPI scores accounting for the difference (p-value = 0.15). Notably, 173/584 patients never received systemic therapy, and 102 of these were expectantly monitored without any local therapy, such as radiotherapy or therapeutic surgery). None of these 102 patients had transformation within the first 36-months of follow up. Finally, we confirm Ginè et al.’s earlier finding that a higher risk FLIPI score confers a higher risk of transformation. (Annals of Oncology, 2006) For each unit increase of FLIPI risk score (e.g., 3 à 4), the probability of histologic transformation at any time point increases 1.72 fold. Moreover, high-risk FLIPI patients (3–5 risk factors) have a 3.3-fold increase in risk of HT (p-value <0.0001).

Conclusions: Patients diagnosed with FL in 2001–2006 received rituximab earlier in their disease and more frequently than those diagnosed in 1998–2000. However, in contrast to our hypothesis, this did not translate to a lower risk of transformation for the 2001– 2006 cohort. The 36-month risk of transformation was lower in patients with lower FLIPI scores. This data supports the clinical decision to expectantly monitor low-risk FLIPI patients.