The Oncologist, Vol. 11, No. 6, 666-673, June 2006; doi:10.1634/theoncologist.11-6-666 © 2006 AlphaMed Press
Liposomal Doxorubicin, Cyclophosphamide, and Etoposide and Antiretroviral Therapy for Patients with AIDS-Related Lymphoma: A Pilot Studya University of Washington School of Medicine, Seattle, Washington, USA; b Ovation Research Group, Highland Park, Illinois, USA, and University of Washington School of Public Health, Seattle, Washington, USA; c Virginia Mason Medical Center, Seattle, Washington, USA Key Words. Non-Hodgkins lymphoma • HIV/AIDS • Antiretroviral therapy • Chemotherapy Correspondence: David M. Aboulafia, M.D., Section of Hematology/Oncology, Virginia Mason Medical Center, 1100 Ninth Avenue, P.O. Box 900 (H14-HEM), Seattle, Washington 98111, USA. Telephone: 206-341-1284; Fax: 206-223-6914; e-mail: hemdma{at}vmmc.org Received November 18, 2005; accepted for publication March 16, 2006.
Purpose. To evaluate in a pilot study the safety and efficacy of liposomal doxorubicin, cyclophosphamide, and etoposide (LACE) when combined with antiretroviral therapy (ART) in patients with AIDS-related lymphoma (ARL). The impact of HIV viral control on therapy and survival was also assessed. Patients and Methods. Between 1994 and 2005, 40 patients at Virginia Mason Medical Center were diagnosed with ARL. Twelve received LACE every 28 days. All patients received intrathecal chemoprophylaxis, ART, and G-CSF. Results. The median patient CD4+ count was 190 cells/µl (range, 20510 cells/µl), and the median HIV viral load (VL) was 61,613 copies/ml (range, <50500,000 copies/ml). Seven patients (58%) had an International Prognostic Index score of 3 or 4. Six patients (50%) were ART-naïve, five were viremic despite ART, and one had an undetectable HIV-1 VL. Nine patients (75%) achieved a complete response (CR), and median overall survival was 107 months. At a median follow-up of 46 months, the recurrence-free survival rate was 50%. Two patients died from relapsed/refractory ARL and one patient achieved a CR with salvage therapy. One CR patient died from complications of pneumonia, and another CR patient died from uncertain causes 5 years after treatment. Grade 3 or 4 neutropenia occurred in 23 of 61 (38%) chemotherapy cycles. Hospitalization was required after 5% of treatment cycles due to neutropenic fever. Conclusion. LACE is an effective and tolerable treatment for ARL. HIV viral control can be maintained in the majority of patients during and after completion of LACE.
HIV-infected individuals are at greater risk of acquiring non-Hodgkins lymphoma (NHL) than their HIV-seronegative counterparts [1, 2]. Recent studies show a decrease in the incidence of AIDS-related lymphomas (ARLs) with the advent of highly active antiretroviral therapy (HAART) [35]. Despite this decrease, ARL accounts for approximately 11% of AIDS-related mortality [6]. Optimal treatment of ARL is poorly defined. Initial regimens in the pre-HAART era mirrored those treatments that were commonly used to treat intermediate and high-grade lymphomas [7]. More complex and aggressive NHL chemotherapy regimens achieved complete remission (CR) in only 20%58% of HIV-infected patients, resulted in severe hematologic toxicity, and contributed to a heightened risk of life-threatening opportunistic infections (OIs) in 20%28% of patients [810]. In an effort to minimize toxicities, the National Institute of Allergy and Infectious Disease AIDS Clinical Trials Group initiated a trial comparing standard-dose with lower-dose m-BACOD (methotrexate, bleomycin, doxorubicin, cyclophosphamide, vincristine, and dexamethasone) chemotherapy for the treatment of ARL. The lesser-dose arm led to similar response rates as the higher-dose arm but caused significantly fewer toxicities [11]. Encouraged by evidence suggesting superior response to administration of chemotherapy via an infusional schedule in acute lymphocytic leukemia, Sparano and colleagues treated 14 patients, of whom 12 had ARL, with a pilot regimen of infusional cyclophosphamide, doxorubicin, and etoposide (CDE), achieving a CR in 71% of patients, although 36% of patients were diagnosed with an OI while on treatment [12]. In the post-HAART era, chemotherapy-associated OIs are greatly diminished [5]. However, potential drugdrug interactions between cytotoxic chemotherapy and concurrent HAART continue to be of concern [13]. In an effort to minimize potential drugdrug interactions, Little and colleagues chose to suspend HAART while patients received dose-adjusted infusional chemotherapy. They achieved a CR rate of 74% among 39 patients, with a median follow-up of 53 months [14]. Doxorubicin is a broad-spectrum antineoplastic agent routinely enlisted to treat a variety of malignancies, including NHL. Adequate dosing of doxorubicin is limited by, among other things, its short- and long-term cardiotoxic side effects. Its rapid clearance from the circulatory system and widespread distribution may contribute to early cardiomyopathy or later-onset ventricular dysfunction [15]. Liposomal-encapsulated derivatives of doxorubicin avoid uptake by the reticuloendothelial system. Consequently, they remain in circulation for longer periods of time, allowing greater distribution in tumors where vessels are abnormally permeable [16]. The longer half-life of polyethylene-coated liposomal formulations of doxorubicin may explain, in part, why they are associated with a lesser risk of neutropenia, anemia, alopecia, and cardiotoxicity [17, 18]. For the treatment of AIDS-related Kaposis sarcoma, liposomal anthracyclines alone produced similar tumor shrinkage as a combination regimen of doxorubicin, bleomycin, and vincristine, but without evidence of cardiotoxicity, neuropathy, significant alopecia, or mucositis [19]. Liposomal formulations of anthracyclines are increasingly being used to treat a variety of solid tumors and lymphoproliferative disorders, including ARL [2022]. Herein, we report the results of a single-institution pilot study to evaluate the safety and efficacy of a novel chemotherapy regimen for ARL consisting of liposomal doxorubicin, cyclophosphamide, and etoposide (LACE), given in conjunction with antiretroviral therapy (ART). By substituting liposomal doxorubicin for doxorubicin and by substituting oral for infusional etoposide as was employed in the CDE regimen of Sparona and colleagues [12], we hoped to achieve a high response rate by preserving the potential benefits of infusional doxorubicin-based chemotherapy but without the downside of central venous line placements, protracted hospitalizations, and use of portable infusion pumps. By substituting oral for i.v. etoposide, we also sought to minimize the inconvenience to the patient of more frequent clinic visits. We were also interested in assessing the impact of HIV viral control on tumor response rates and patient survival.
Eligibility and Staging Requirements for enrollment were an age of at least 18 years, HIV-positive serology, a Karnofsky performance status of 40%, and biopsy-proven intermediate- or high-grade ARL. Exclusion criteria included prior chemotherapy or radiation therapy, a history of impaired cardiac status, active and life-threatening infections, pregnancy or nursing, and a diagnosis of primary central nervous system lymphoma. Initial staging studies included a computerized tomography (CT) scan of the chest, abdomen, and pelvis; 12-lead electrocardiogram; cardiac echocardiogram to assess left ventricular ejection fraction (LVEF); posterior iliac crest bone marrow aspirate and biopsy; and lumbar puncture for cytologic analysis of the cerebrospinal fluid. Laboratory studies included serial assessment of CD4+ T-lymphocyte cell count, HIV-1 viral RNA load (VL), ß2 microglobulin, blood chemistries including lactate dehydrogenase (LDH), and complete blood count. We used the Ann Arbor staging system for designating lymphoma stage [23]. Gallium and positron emission tomography (PET) scans were not routinely used for staging patients, although several patients did undergo either gadolinium or PET scan assessments off protocol. The LACE protocol was reviewed and approved by the Virginia Mason Medical Center (VMMC) Institutional Review Board. All patients provided informed consent before agreeing to protocol therapy.
Treatment and Evaluation
Responding patients received five to eight cycles of chemotherapy. Doses were modified or delayed if the absolute neutrophil count (ANC) was <750/mm3 or platelet count was <50,000/mm3. After a 1-week delay, chemotherapy was resumed at the original doses if the ANC rose above 750/mm3 and platelets increased to >50,000/mm3. If the ANC or platelet count remained below these values after 1 week, therapy was delayed for a second week; in these patients, once the ANC and platelet counts returned to >750/mm3 and 50,000/mm3, respectively, cyclophosphamide and etoposide chemotherapy was resumed at 75% of the original dosing. All known sites of disease were radiologically reassessed after every two cycles of LACE, at therapy termination, every 3 months for the first year following treatment, and at 6-month intervals thereafter. Standard international workshop criteria were used to evaluate response to treatment [25]. A CR was defined as a resolution of all detectible disease, without development of new lesions. Partial response (PR) was defined as
Statistics
Patient Characteristics Patient demographics and ARL prognostic factors are reviewed in Table 1
Treatment and Outcome Of the 12 patients enrolled, five (42%) received the intended number of chemotherapy cycles. Reasons for early termination from the study included death (one patient), progressive disease (PD) or stable disease (two patients), and patient noncompliance (two patients). Table 2
Toxicity Aspects of chemotherapy-associated toxicity are reviewed in Table 3
Impact of HIV VL on Patient Response All patients received ART in conjunction with chemotherapy (Table 3 In total, 3 of the 12 patients responded poorly to LACE (patient 1, 10, and 12). Patient 1 had uncontrolled viremia at initiation of chemotherapy and died 2 months later of PD. Despite obtaining an undetectable HIV VL, patient 10 died of relapsed ARL 2 months after unsuccessful high-dose chemotherapy and stem cell reinfusion. Patient 12 averaged an HIV VL of approximately 7,000 copies/ml and achieved a CR after high-dose cyclophosphamide salvage chemotherapy. All four patients (patient 46 and 9) with undetectable HIV VLs at trial termination achieved a CR with LACE.
At the time this pilot trial began, the capacity for potential drugdrug interactions between chemotherapy regimens and ART was poorly understood. In an effort to test the pharmacokinetic interactions of CHOP with HAART, the AMC demonstrated that the protease inhibitor (PI) indinavir decreased cyclophosphamide clearance by 1.5-fold [29]. Despite a somewhat higher incidence of myelosuppression and autonomic neurotoxicity, the authors concluded that HAART, given concurrently with CHOP, is both safe and tolerable but warned that trials using higher doses of cyclophosphamide should be monitored carefully. Antinori and colleagues treated ARL patients with concurrent HAART and various chemotherapies and were able to achieve a higher CR rate and an acceptable toxicity profile [30]. Altered ART pharmacokinetics or poor antiretroviral drug compliance during chemotherapy may lead to increased HIV resistance. It is thus reassuring that a recent study was able to show that HIV virologic response could be maintained in the majority of ARL patients who received cyclophosphamide and anthracycline-based treatments [13]. Despite vagaries of drugdrug interactions and ART compliance, LACE given concurrently with ART did not appear to be associated with undue toxicity. This studys hematologic toxicity profile appears comparable with that of the NCI continuous-infusion etoposide, prednisone, vincristine, cyclophosphamide, hydroxydaunorubicin (EPOCH) study with HAART suspension. LACE participants experienced neutropenia of <1,000 cells/mm3 after 38% of cycles, and grade 3 or 4 anemia and thrombocytopenia in 3% and 5% of cycles, respectively, versus neutropenia of <500 cells/mm3 after 30% of cycles and grade 3 or 4 anemia and thrombocytopenia in 17% and 21% of cycles, respectively, in the NCI study [14]. Patients in the NCI study experienced neutropenic fever in 13% of cycles versus 5% in LACE patients. No undue episodes of oral thrush were reported by study participants while they received prophylactic fluconazole. Although fluconazole has been proven to inhibit the bioactivation of cyclophosphamide via cytochrome P-450-mediated pathways, studies show that this interaction is of minor significance and is not considered clinically relevant [31]. Furthermore, no significant cardiac toxicity was identified in this study, which is consistent with the experiences of other investigators who have used liposomal doxorubicin in several other clinical venues [32]. Finally, patients enrolled in this pilot study did not routinely receive corticosteroids. Previous studies by Sparona and colleagues using CDE chemotherapy indicated that steroids were not essential for the treatment of ARL [12, 33]. It was our goal to find a convenient and effective regimen that would limit risks for additional infections (including reactivation of chronic viral infections) and hyperglycemia, and would not exacerbate manic behavior or depression. Although the response rate to LACE in this pilot study is encouraging, the study has several obvious limitations. In addition to a small sample size and varied ART regimens, several patients possessed comorbid psychiatric issues undoubtedly affecting their ART and chemotherapy compliance as well as their choices in health care maintenance. Three of the participants had a history of severe and chronic depression (patients 79). In particular, patient 8 was diagnosed with a dissociative disorder and amphetamine addiction and refused his last cycle of chemotherapy. He was lost to follow-up and succumbed from pneumonia after achieving a CR. Patient 4 had a history of active methamphetamine abuse that led to erratic behaviors, missed appointments, and delays in treatment. Patient 5 was emotionally labile and often irreconcilably agitated. He missed several chemotherapy appointments that led to delays in his treatment. An additional factor that led to the modest number of patients accrued to this study was that in 1999 the VMMC was accepted into the AMC. Because of a competing ARL protocol that randomized patients to CHOP or CHOP-R, only patients who were ineligible or who declined participation in AMC Study 010 were offered participation in the LACE pilot study. Although a minority of ARL patients seen at the VMMC were entered into this protocol, it is worth noting that the preponderance of patients with advanced NHL, low CD4+ cell count and elevated HIV VL, elevated LDH, and high International Prognostic Index score are indicators that this was not a particularly favorable NHL group to treat [27]. Although not included in this studys statistics, three additional patients with AIDS-related Hodgkins disease were also treated with the LACE regimen. Two of those patients achieved a CR, one of which was still in remission 53 months postchemotherapy. The other patient, with an HIV VL of 188,000 at trial termination, was the only patient who relapsed and died of recurrent Hodgkins lymphoma after achieving a CR with LACE chemotherapy. The third patient, who did not respond to the LACE regimen, achieved a CR after ABVD salvage chemotherapy and remained in remission 48 months post-ABVD (Doxorubicin, Bleomycin, Vinblastine, and Dacorbazine). Recently, Bower and colleagues investigated the effect of PI-sparing ART regimens versus PI-containing regimens among 46 Westminster and Chelsea ARL patients treated with CDE chemotherapy [34]. The day 10 and day 14 (p = .025) neutrophil counts were significantly lower in patients receiving PIs, although there was no difference in the number of days of G-CSF administered between the groups. Unfortunately, our study is too small to make specific inferences regarding the relative risk of concurrent PI- or non-PI-containing regimens on hematologic toxicity. Importantly, the British investigators were not able to show differences in response rates, relapse-free survival, or disease-free survival between the two groups. Nonetheless, their finding of greater hematologic toxicity among patients treated with chemotherapy and PI-based ART underscores the importance of performing careful pharmacokinetic studies on ARL patients who enroll in trials that incorporate chemotherapy concurrently with ART. The NCI-sponsored AMC recently evaluated the efficacy of CHOP and CHOP-R in conjunction with HAART for the treatment of ARL. Despite a superior patient response compared with CHOP alone (58% vs. 47%; p = .147), CHOP-R was associated with significantly higher rate of treatment-related infectious death in ARL patients [35]. Compared with continuous-infusion regimens like dose-adjusted EPOCH and CDE [12, 14, 33, 34, 36], out-patient administration of LACE may be more convenient. Treatments can be given via a peripheral vein. Central line catheter placement is not routinely necessary. Patients need not remain in the hospital for long time periods. Attempts at finding an outpatient-based treatment with response rates and tolerability that are comparable with those of continuous-infusion EPOCH are challenging. A nonpegylated liposomal doxorubicin-based CHOP regimen was recently evaluated in patients with newly diagnosed ARL [22]. Nonpegylated liposomal doxorubicin was given in doses of 40, 50, 60, and 80 mg/m2, with fixed doses of cyclophosphamide, vincristine, and prednisone every 21 days. There was no dose-limiting toxicity, although myelosuppression was the most frequent toxicity. The overall CR rate was 75%, and the median duration of CR was 15.6 months. In conclusion, LACE appears to be a safe and well-tolerated treatment for ARL. Its favorable toxicity profile and options for outpatient administration on a once-monthly basis make it a convenient regimen for patients with ARL.
The authors indicate no potential conflicts of interest.
We wish to acknowledge the expert and invaluable clinical care of Denise Bundow, ARNP, and Ms. Bonnie Marston for manuscript preparation. Presented as a poster presentation at the 9th International Conference on Malignancies in AIDS and Other Immunodeficiencies: Basic, Epidemiologic and Clinical Research, Bethesda, Maryland, September 2727, 2005, and supported through the Medical Research Training program at the University of Washington School of Medicine.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||