Ritonavir Based Highly Active Retroviral Therapy (HAART) Correlates with Early Neurotoxicity When Combined with ABVD Treated HIV Associated Hodgkin Lymphoma but Not Non-Hodgkin Lymphoma. A Retrospective Study.

Hodgkin (HL) and non-Hodgkin lymphoma (NHL) are the most common non-HIV-defining and HIV-defining malignancies, respectively¹. Patients (pts) treated for lymphoma concomitantly with highly active retroviral therapy (HAART) have much better response rates and overall survival². However, few studies have assessed the impact of HAART on the adverse events of chemotherapy. The Stanford V regimen for HL has a 48 % increase in neurotoxicity and neutropenia when taken concomitantly with HAART compared to non-HIV patients³. Case reports exist associating gastrointestinal and neurotoxicity with vinblastine (less so vincristine) and ritonavir-based HIV therapy. Unfortunately, the frequency of this association has yet to be shown. We retrospectively analyzed pts at our institution with lymphomas and HIV and the relationship between HAART and chemotherapy adverse events (AE) in an attempt to minimize future chemotherapy-related complications.

We retrospectively identified HIV infected patients treated for lymphoma (2002-2008) at Stroger Hospital of Cook County by computer database search using ICD9 and pharmacy codes and then confirmed by medical record review. The adverse events during chemotherapy were assessed by chart review and graded per the NCI Common Terminology Criteria for Adverse Events. Those pts where HAART or chemotherapy regimen could not be confirmed were excluded from the study. Patient, disease, and the HIV characteristics were assessed.

A non-parametric Fisher's exact test was used to examine the difference in proportion of patients that developed an adverse event in the pts with a particular HAART medication and the ones without.

Twenty-five HIV infected pts with HL and 46 pts with NHL were identified. Identification of HAART administered during chemotherapy was found in 23/25 (92%) pts with HL and 22/46 (48%) pts with NHL. Patients diagnosed with HL were 87% male with a median age of 43. Mixed cellularity (26%) and nodular sclerosis (26%) pathologies were the most common with 60% (14/23 pts) presenting as stage III/IV disease. The median CD4 count at diagnosis was 174 cell/ml with a range of 26–341. The most common HAART regimen was efavirenz/emtricitabine/tenofovir (17%), though ritonavir-based regimens accounted for 48% of all HIV treatments. Twenty-six percent (6/23 pts) had neurotoxicity (13% G3, 4% G2, 8% G1) and among those with any neurotoxicity,100% were taking ritonavir. Each of the 3 pts with G3 neurotoxicity, and the patient with G3 constipation experienced their AE after just 1 to 2 doses of vinblastine (before cycle #2). Each pt with G3 neurotoxicity, vinblastine was held with only slight improvement in symptoms. A statistically significant difference in the proportion of patients that developed neurotoxic effects after initiation of vinblastine was observed in the ritonavir vs. non-ritonavir based HAART therapy groups (55% vs. 0%, p<0.01). The most common non-hematological side effects were nausea and vomiting (29%) and neuropathy (28%; 13% G3 and 25%G1 and 2). Twenty-nine percent of the pts had G4 neutropenia. Two deaths occurred during induction therapy secondary to CMV pneumonia and cryptococcal infection.

In contrast, no correlation was found between any HAART and AE in any vincristine-based chemotherapy regimen used to treat NHL. The most common diagnosis was diffuse large B cell lymphoma (63%). All 22 pts received CHOPR (63%), HYPERCVAD (27%), or CHOP (9%). Thirty-two percent (7/22 pts) were placed on a ritonavir based HAART, but no G3/4 neurotoxicity were observed. No G4 complications were seen aside from neutropenia (18%).

Ritonavir-based HIV therapy in conjunction with ABVD (vinblastine based) in the treatment of HL was both temporally and statistically associated with serious, sometimes irreversible neurologic toxicity after just one dose of vinblastine. This same relationship was not seen in vincristine-based therapies for NHL. These data suggest that the CYP3A4 enzyme inhibition caused by ritonavir leads to severe vinblastine-associated adverse events and clinicians should consider non-ritonavir based HIV therapy during ABVD treatment for HL.

No relevant conflicts of interest to declare.