Gemcitabine and Other New Cytotoxic Drugs: Will Any Find Their Way Into Primary Therapy?
- First Online:
- Cite this article as:
- Dougherty, D.W. & Friedberg, J.W. Curr Hematol Malig Rep (2010) 5: 148. doi:10.1007/s11899-010-0054-x
- 112 Views
Primary treatment for classic Hodgkin lymphoma (HL) remains highly effective with chemotherapy alone or combined-modality therapy. The limitations of therapy have been related to toxicity and efficacy in subsets of patients. The introduction of a number of new and novel cytotoxic agents has provided opportunities for investigating their use in the treatment of HL. This article briefly reviews current primary treatment strategies for HL and examines the existing data for both new cytotoxic agents and other selected novel agents in the treatment of HL.
Classic Hodgkin lymphoma (HL) is a relatively rare lymphoid neoplasm with an estimated incidence of 8510 cases and 1290 deaths in the United States in 2009 . Overall, HL remains one of the most curable of all cancers. Current therapies for early-stage HL (stages IA and IIA) include short courses of combination chemotherapy—usually doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD)—followed by involved-field radiotherapy, resulting in cure rates of 90% to 95% . Thus, the major challenge in the treatment of early-stage HL has been to maintain these excellent treatment results while reducing short-term and long-term toxicities. Patients with advanced-stage HL (stages III–IV) are generally treated with six to eight cycles of combination chemotherapy, with ABVD the regimen currently preferred in the United States. Long-term survival rates for patients with advanced disease remain on the order of 50% to 60%, suggesting the need for better therapies and prognostic mechanisms for these patients [3–5]. The advent of new cytotoxic agents, such as gemcitabine, has prompted investigation into their utility in the treatment of HL, both for initial treatment and in the relapse setting. Additionally, a number of other novel agents that demonstrate cytotoxicity have been introduced and may be used to further improve upon the standard treatments for HL. This article briefly reviews current primary treatment strategies for HL and examines the existing data for both new cytotoxic agents and other selected novel agents in the treatment of HL.
Current Treatment Approaches
The treatment of early-stage HL has evolved considerably over the past 30 years. Initial strategies involving radiation therapy (RT) alone have given way to the current standard of care of combined-modality therapy with chemotherapy and RT. Combined-modality regimens have achieved excellent outcomes, with complete response (CR) rates greater than 90% and superior freedom from treatment failure (FFTF) rates of 88% at 7 years, compared with 67% for RT alone [6, 7]. Given these results, a greater emphasis has been placed on establishing strategies to maintain these outcomes while decreasing the long-term toxicities of treatment, most of which have been related to late complications of RT, although the overlapping toxicity of doxorubicin and bleomycin with RT has also been of concern. Recent studies have attempted to lessen toxicity by reducing the dosage and field size of RT, decreasing the number of cycles of chemotherapy, and eliminating radiation altogether [8••, 9, 10].
The development of the MOPP (mechlorethamine, oncovorin, procarbazine, prednisone) combination chemotherapy regimen in the 1960 s represented a major advance in the treatment of patients with stage III–IV HL and offered a cure to a significant proportion of these patients. Long-term follow-up suggesting 54% relapse-free survival and 68% disease-free survival at 10 years with MOPP was tempered by the findings of significant acute and long-term toxicity, including an increased incidence of acute myeloid leukemia and sterility . Efforts to develop alternative therapies led to the development of the non–cross-resistant ABVD regimen by Bonadonna and colleagues . Subsequent randomized trials proved ABVD to be more effective and less toxic than MOPP. Further studies showed that alternating (MOPP-ABVD) and hybrid (MOPP-ABV) regimens were no more effective than ABVD and had greater toxicities [3, 5, 13]. These studies established ABVD as the standard of care in the United States, and long-term follow-up of high-risk patients treated with ABVD shows a 14-year failure-free survival (FFS) of 47% and overall survival (OS) of 59% . Additionally, most series have noted substantial acute and long-term cardiopulmonary toxicities with ABVD, as well as a lower, though present, risk of secondary malignancy and myelodysplastic syndrome.
Further efforts have been made to improve upon standard therapy by increasing the dose-intensity of chemotherapy through reduction in treatment duration, reducing the cumulative doses of drugs associated with long-term toxicities, and reducing the extent of radiation. One such regimen under investigation is Stanford V, which reduces the cumulative doses of bleomycin, Adriamycin, and mustard, and omits procarbazine . A recent US Intergroup trial (E2496) randomized patients to either ABVD or Stanford V; the results of this trial are not yet available.
The German Hodgkin’s Study Group (GHSG) used the principles of dose escalation and time intensification in developing the BEACOPP regimen (bleomycin, etoposide, Adriamycin, cyclophosphamide, oncovorin, procarbazine, and prednisone), which is delivered on a 21-day cycle . An escalated BEACOPP variant, using increased doses of cyclophosphamide, Adriamycin, and etoposide with granulocyte colony-stimulating factor (G-CSF) support, was compared with standard BEACOPP and COPP-ABVD in the three-armed, randomized HD9 trial. Recently published long-term follow-up data from HD9 support the results seen in the initial analysis. At 10 years, FFTF rates were 64%, 70%, and 82% respectively for patients treated with COPP-ABVD, standard BEACOPP, and escalated BEACOPP; the OS rates were 75%, 80%, and 86% [16••]. Significant differences were seen in the number of patients dying from HL (11.5% vs 8.1% vs 2.8%) for COPP-ABVD, standard BEACOPP, and escalated BEACOPP and in actual overall mortality (24% vs. 19% vs. 12%).
Despite these advances in primary therapies, opportunities remain to improve outcomes among various subsets of HL patients. Elderly patients continue to face a poorer prognosis than younger patients with HL . Patients presenting with a high International Prognostic Score (IPS) or those with positive early interim positron emission tomography (PET) remain at increased risk for treatment failure [18–20]. Though risk-adapted and response-adapted treatment strategies are important approaches in improving outcomes in these patients, new and novel agents may also hold promise in the initial treatment of HL.
Newer Cytotoxic and Novel Agents
Therapy for Relapsed or Refractory Hodgkin Lymphoma
Selected prospective trials of gemcitabine in Hodgkin lymphoma
ORR (CR), %
Relapsed and refractory HL
Santoro et al. 
Median response duration 6.7 mo
Venkatesh et al. 
Median time to progression 6.4 mo
Zinzani et al. 
4 of 6 response duration >12 mo
Baetz et al. 
All patients with successful stem cell collection
Bartlett et al. [27•]
52% 4-y event-free survival in transplant-naive patients
Santoro et al. 
78 of 79 patients with successful stem cell collection
Cole et al. [29•]
Gemcitabine & vinorelbine
Median age 17.7 y (range, 10.7–29.4)
Oki et al. 
Gemcitabine & rituximab
Median response duration 3.7 mo; 52% ORR in CD20-negative HL
Mendler et al. 
Gemcitabine & bortezomib
Combination less active and more toxic than gemcitabine alone
Primary therapy for HL
Friedberg et al. 
Trial stopped prematurely with 42% incidence of pneumonitis
Bredenfeld et al. 
Trial stopped prematurely with 30% incidence of pneumonitis
Straus et al. [35•]
CR rate lower and PFS shorter than ABVD
The National Cancer Institute of Canada initially reported results from a phase 2 trial of gemcitabine, dexamethasone, and cisplatin (GDP) in 23 patients with relapsed or refractory HL . Four patients (17%) achieved CR, 12 (52%) achieved partial response (PR), and all patients had successful stem cell collection and underwent transplantation. In a retrospective analysis, 68 patients with refractory or relapsed HL received either GDP or mini-BEAM (carmustine, etoposide, cytarabine, and melphalan) followed by autologous hematopoietic stem cell transplantation (ASCT) . Nonsignificant differences were noted in ORR (62% for GDP vs 68% for mini-BEAM) and stem cell collection (97% for GDP vs 82% for mini-BEAM), but progression-free survival at 18 months was significantly better for the GDP group (74%) than for the mini-BEAM group (35%).
The Cancer and Leukemia Group B (CALGB) studied the combination regimen GVD (gemcitabine, vinorelbine, and liposomal doxorubicin) as a salvage regimen in a cohort of 91 patients with relapsed HL, including patients with relapse after SCT [27•]. In this phase 1/2 trial, there was one treatment-related death due to alveolar hemorrhage in a patient who had previously undergone SCT; grade 3/4 leukopenia, neutropenia, and mucositis were more common in transplant-naive patients. Overall, GVD was well tolerated and led to an ORR of 70% with 19% complete remissions.
Similarly promising results were reported by Santoro et al.  in a separate phase 2 trial of 91 patients treated with the IGEV (ifosfamide, gemcitabine, vinorelbine, and prednisolone) regimen. IGEV was well tolerated, with low rates of grade 3/4 neutropenia (5.7%) and thrombocytopenia (4.8%) and only one episode of grade 4 mucositis. The ORR was 81%, with a CR rate of 54% and adequate stem cell collection in 78 of 79 patients in which it was attempted. The results with both GVD and IGEV suggest that high activity with an acceptable toxicity profile make these regimens reasonable choices for salvage induction therapy.
A recent phase 2 study from the Children’s Oncology Group assessed the efficacy and toxicity of weekly gemcitabine and vinorelbine (GV) in 30 pediatric patients with heavily pretreated relapsed/refractory HL [29•]. The median age of patients in this study was 17.7 years, with a range of 10.7 to 29.4 years. Of 25 evaluable patients, 19 had measurable responses, for an ORR of 76% and a 24% CR rate. Toxicity was tolerable and consisted primarily of reversible myelosuppression. On the basis of these results, further evaluation of GV for patients with HL seems warranted.
Others have examined gemcitabine in combination with other novel agents. Rituximab, the monoclonal antibody against CD20, has shown activity as a single agent in heavily pretreated patients with HL . In light of this finding, a subsequent study evaluated rituximab combined with gemcitabine in 33 patients with refractory or relapsed HL . Responses were observed in 48% of patients regardless of CD20 positivity, and grade 3/4 toxicities included only neutropenia (36%) and thrombocytopenia (15%). Based on the ability of bortezomib, a proteasome inhibitor, to inhibit NF-κB, this agent was examined in combination with gemcitabine in 18 patients with relapsed or refractory HL . The ORR was only 22%, and reversible liver toxicity was noted in three patients, with one patient being removed from the study and two having doses held due to transaminase elevation, calling attention to the importance of prospective safety studies of novel combinations. Though the combination of bortezomib and gemcitabine should not be pursued in treating HL, the results with the combination of rituximab and gemcitabine, with a favorable safety profile and ease of outpatient administration, may warrant further investigation with other active agents.
Primary Therapy for Hodgkin Lymphoma
The effectiveness of gemcitabine in the treatment of HL patients with relapsed or refractory disease, along with relatively minimal associated toxicity, has prompted its study in patients with newly diagnosed HL. In an effort to improve upon the efficacy of ABVD, we substituted gemcitabine for dacarbazine in the standard regimen to form ABVG (doxorubicin, bleomycin, vinblastine, and gemcitabine) . Twelve patients with advanced-stage de novo HL were enrolled and the maximally tolerated dose of gemcitabine was determined to be 800 mg/m2 in this combination. The trial was stopped prematurely because of an unexpectedly high incidence and severity of pneumonitis, with five patients (42%) experiencing clinically significant pulmonary toxicity in cycles 4 to 6. The response rate in the study was 100%; five patients experienced disease progression during a median follow-up of 1 year. The increased incidence and severity of pulmonary toxicity in this regimen (as compared with regimens not containing gemcitabine) suggested a possible interaction between gemcitabine and bleomycin, so the authors recommended that the bleomycin/gemcitabine combination not be pursued for de novo HL.
The GHSG examined a less leukemogenic BEACOPP regimen that substituted gemcitabine for etoposide for patients with advanced-stage de novo HL . The phase 1/2 study evaluated BAGCOPP (bleomycin, doxorubicin, gemcitabine, cyclophosphamide, vincristine, procarbazine, and prednisone) in 27 patients. Gemcitabine was given at a starting dose of 800 mg/m2 on days 1 and 4 of each cycle. Because of higher-than-expected hematotoxicity, gemcitabine on day 4 was omitted. Of the 27 patients treated, 8 experienced significant pulmonary toxicity and the study was stopped early. Pneumonitis was seen in six patients, 1 of whom died while in CR, with underlying pulmonary fibrosis and Pneumocystis carinii pneumonia. Another two patients showed a significantly reduced diffusing capacity of the lung for carbon monoxide (DLCO), but all patients except the one who died recovered from pulmonary toxicity. Despite a response rate of 93% (25 of 27) and with all responders in continued CR at a median follow-up of 27 months, the pulmonary toxicity was unacceptable. The authors hypothesized that this toxicity was related to the combination of gemcitabine and bleomycin.
In an attempt to mitigate bleomycin pulmonary toxicity with ABVD, the CALGB conducted a phase 2 trial of AVG (doxorubicin, vinblastine, and gemcitabine) for the initial treatment of stage I and stage II nonbulky HL [35•]. The primary end point of CR plus unconfirmed CR (CRu) was observed in only 67 (67.7%) of 99 patients treated, and the estimated progression-free survival was 81% at 1 year and 71% at 2 years. With a CR rate that was lower and a progression-free survival shorter than what would be expected with standard ABVD, this regimen does not appear to warrant further evaluation for initial treatment of HL.
Vinorelbine is a semisynthetic vinca alkaloid that is more selective than other vinca alkaloids for interfering with polymerization of mitotic microtubules . When used as a single agent in heavily pretreated patients with relapsed or refractory HL, response rates as high as 50% have been seen . Much of the study of vinorelbine in HL has been in the relapsed/refractory setting, and its nonoverlapping mechanism of action with gemcitabine has made this combination of particular value. The response rates and acceptable toxicity profiles of combination regimens including vinorelbine and gemcitabine (GVD, IGEV, and GV) have allowed their use as effective salvage therapies.
Before the development of the previously mentioned gemcitabine and vinorelbine combination regimens, Ferme et al.  studied a vinorelbine-containing regimen in 100 patients with relapsed or refractory HL. The MINE regimen (mitoguazone, ifosfamide, vinorelbine, and etoposide) produced an ORR of 75%, but three patients died of MINE-related complications. Subsequently, a salvage regimen of high-dose ifosfamide (3000 mg/m2 per day on days 1–4) and vinorelbine (25 mg/m2 on days 1 and 5) was examined in 47 patients . This regimen showed an ORR of 83%, and grade 3/4 neutropenia was seen in 65%, suggesting that a regimen of high-dose ifosfamide and vinorelbine was both active and well tolerated.
To date, the only published attempt to evaluate the response to vinorelbine in the initial treatment of HL was by Benchekroun et al. . Thirty-two patients with advanced-stage HL received four weekly doses of vinorelbine prior to MOPP/ABVD; PR was achieved in 27 (90%) of 30 evaluable patients. On this basis, it is unclear whether vinorelbine provides any clear benefit in the primary treatment setting.
Rituximab, a monoclonal antibody against CD20, was studied initially in heavily pretreated HL patients. Though only 25% to 30% of the malignant Hodgkin and Reed-Sternberg (HRS) cells in classic HL express CD20, rituximab has been shown to have activity both as a single agent (22% ORR) and in combination with gemcitabine (48% ORR) in patients with recurrent or refractory HL [30, 31]. Although the mechanism of this effect is not entirely delineated, it is known that CD20-positive B cells make up much of the microenvironment surrounding HRS cells and may promote the survival of HRS cells by providing ligands needed to sustain HRS cells. The depletion of these reactive lymphocytes, therefore, may render the HL cells more sensitive to chemotherapy. Further, it has been shown recently that the HL stem cell may express the CD20 antigen, which would provide another possible target for rituximab .
Based upon these data, the same M.D. Anderson group examined the addition of rituximab to ABVD (RABVD) in newly diagnosed patients with advanced-stage HL, and the final results of this phase 2 study were recently reported [42•]. Rituximab was administered at 375 mg/m2 weekly for 6 weeks beginning either concurrently with the first dose of ABVD or 3 weeks before the first dose of ABVD. The authors also conducted a retrospective analysis of their own institutional data for patients with advanced HL treated at their center with ABVD from February 1996 to May 2006. Of the 104 evaluable patients, 78% had either stage III or stage IV disease and 37% had IPS greater than 2. With a median follow-up of 5 years, the projected event-free survival (EFS) for RABVD is 87%, which was significantly better than the institutional 5-year EFS for ABVD patients (66%, P = 0.0036). Improvement in EFS was observed with RABVD in patients with IPS 0 to 2 (89% vs 71% with ABVD, P = 0.0248) and in those with IPS greater than 2 (80% vs 55% with ABVD, P = 0.0532). These data serve as the basis for an ongoing multicenter randomized trial comparing ABVD with RABVD in newly diagnosed patients with advanced-stage classic HL with IPS >2. The long-term results of this trial may help to further clarify the role of rituximab in the initial treatment of HL, as will the results of a new GHSG study in which patients with positive early interim PET receive rituximab in addition to BEACOPP.
Bendamustine is a bifunctional alkylating agent developed in the German Democratic Republic (GDR) in 1963 and recently approved by the US Food and Drug Administration for the treatment of chronic lymphocytic leukemia (CLL) and indolent B-cell non-Hodgkin lymphomas (NHL). Despite growing evidence of its utility in these settings, there are few data to support the use of bendamustine in HL. Based upon preliminary data showing efficacy of a bendamustine-containing regimen in HL patients refractory to primary COPP, physicians in the GDR tested the DBVCR B (daunorubicin, bleomycin, vincristine, and bendamustine) regimen in first-line therapy for patients with advanced HL . Patients were randomized to receive either alternating DBVCR B and COPP or COPP alone. With 40 patients in each treatment arm, no significant difference was seen in the remission rate (65% with the regimen including bendamustine vs 80% with COPP alone).
Moskowitz et al.  have recently reported interim results of an ongoing phase 2 clinical trial evaluating the activity of single-agent bendamustine in patients with relapsed or refractory HL following ASCT or nonmyeloablative allogeneic SCT, or in patients who were ineligible for transplantation. To date, 18 of a planned 37 patients have been enrolled and received bendamustine (120 mg/m2) for 2 consecutive days every 28 days. Of 16 evaluable patients, the ORR was 75%, with 38% CR and 38% PR, and toxicity was largely limited to neutropenia, thrombocytopenia, and nausea. Though preliminary, these data suggest efficacy and tolerability of bendamustine in heavily pretreated HL patients, and further study of this agent may be warranted.
Lenalidomide, a novel immunomodulatory drug, is an approved antineoplastic therapy for multiple myeloma and myelodysplastic syndrome. It has shown promising clinical activity and a manageable toxicity profile in a number of B-cell malignancies. Multiple mechanisms of action have been identified for lenalidomide, including direct induction of apoptosis in tumor cells, alteration of the tumor microenvironment, antiangiogenic effects, modification of cytokine profiles, and modulation of immune (T and NK) cells. Based upon these observations, lenalidomide has been examined in the setting of relapsed or refractory HL.
Results of two phase 2 trials with lenalidomide have been reported in relapsed or refractory classic HL. Preliminary results for the first study included 14 evaluable patients, 10 of whom had previously undergone ASCT; the ORR was 14.3%, with PR seen in two patients . Final results of the second trial were recently presented . This trial included 35 patients, who received a regimen of 25 mg per day of lenalidomide on days 1 to 21 of a 28-day cycle. Six clinical responses were noted (1 CR, 5 PR), for an ORR of 17%, and six more patients had stable disease (SD) for longer than 6 months, giving an overall cytostatic response rate (CR + PR + SD) of 34%. Grade 3/4 toxicities were most commonly hematologic, and no tumor lysis syndrome or tumor flare reactions were observed.
In contrast, a recent case series described tumor flare reactions in three patients receiving lenalidomide and methylprednisolone for treatment of refractory HL . These reactions mimicked early tumor progression, but the authors reported that the signs and symptoms resolved upon treatment with anti-inflammatory agents and never recurred. All three patients attained an objective response.
Though these single-agent response rates in relapsed or refractory HL are not as high as rates for other agents, lenalidomide may be explored in combinations or as “maintenance” therapy, given its manageable toxicity profile and oral administration.
Histone Deacetylase Inhibitors
As understanding of the role of epigenetic modulation of gene expression in the pathobiology of cancer has evolved, much interest has been focused on histone deacetylase (HDAC) inhibitors and, more broadly, on deacetylase (DAC) inhibitors as anticancer agents. Epigenetics encompasses changes in genetic expression and cellular phenotype without alterations in DNA sequences. Such changes in the cellular epigenetic environment play a role in tumor formation, progression, and resistance to treatment [48, 49].
To date, more than 15 HDAC inhibitors have been examined in preclinical and early clinical studies. Vorinostat (SAHA), approved for the treatment of refractory cutaneous T-cell lymphoma, has shown activity in HL cell lines and was tested in a phase 2 trial with an ORR of 4% (1 PR in 25 patients) . An oral isotype-selective HDAC, MGCD0103, was tested in a total of 33 patients with relapsed or refractory HL . Patients received either 110 mg or 85 mg three times per week on a 4-week cycle. Among the 21 evaluable patients in the 110-mg cohort, the ORR was 38%, with 2 CR and 6 PR. Of 10 patients in the 85-mg cohort, 5 were evaluable, and all patients had tumor reductions of 30% or more. Toxicity was acceptable, with 20% of patients in the 85-mg cohort and 39% of patients in the 110-mg cohort experiencing grade 3/4 toxicities.
Panobinostat (LBH589), a potent oral pandeacetylase inhibitor, has also been studied in patients with relapsed or refractory HL. A phase 1/2 trial of panobinostat in patients with advanced hematologic malignancies included 31 patients with HL, 28 of whom were evaluable for response assessment . Patients received panobinostat on a Monday/Wednesday/Friday schedule, with doses ranging from 30 to 60 mg every week or 45 to 60 mg every other week, and responses were assessed by both PET and CT. The overall response rate was 68% by PET and 42% by CT, with the most common grade 3/4 toxicities being thrombocytopenia (71.4%), neutropenia (28.6%), fatigue (18.4%), and anemia (6.1%).
Interim results of a two-stage phase 2 study of panobinostat in patients with relapsed or refractory HL after high-dose chemotherapy and ASCT have recently been reported . Of 61 patients enrolled, 53 were evaluable for response. The ORR was 21%, including one CR and 10 PR, and 31 patients had SD. The disease control rate (CR + PR + SD) was 79%. Grade 3/4 toxicities included thrombocytopenia (77%), anemia (20%), and neutropenia (16%).
The results of these studies with panobinostat demonstrate encouraging clinical activity and a tolerable adverse-effect profile. They serve as the foundation for a pivotal international trial that is now ongoing.
Antibody-Drug Conjugates: SGN-35
Monoclonal antibodies such as rituximab, cetuximab, bevacizumab, and trastuzumab have demonstrated considerable utility in the treatment of cancer. Despite their activity, these antibodies are rarely curative. Much attention has been given to enhancing their activity by linking cytotoxic drugs to the antibody, creating antibody-drug conjugates (ADC) capable of site-selective drug delivery. SGN-35 (brentuximab vedotin) is an ADC that delivers the antitubulin agent monomethyl auristatin E (MMAE) to CD30-positive HL cells by binding specifically to CD30 on the cell surface and releasing MMAE. A multicenter phase 1 dose-escalation study was conducted in 45 patients with refractory or relapsed CD30-positive hematologic malignancies, 42 of whom had HL [54•]. The maximally tolerated dose was 1.8 mg/kg every 3 weeks and the most common adverse events were fatigue, fever, nausea, and diarrhea. Among 28 evaluable patients treated with doses greater than 1.2 mg/kg, the ORR was 46%, with a CR rate of 25%. Additionally, 75% of patients with B symptoms at baseline reported resolution with treatment. These remarkable results formed the basis of a recently completed pivotal study of SGN-35 in patients with relapsed HL after ASCT. A phase 3 trial evaluating SGN-35 as a “relapse-prevention” agent (used as maintenance therapy after ASCT) for HL is also under way. Given these promising results, studies of SGN-35 as part of primary therapy for HL or following such therapy are warranted.
Despite advances in the primary treatment of HL, opportunities remain to improve outcomes among various subsets of patients and to decrease long-term treatment-related toxicities. The discovery of a number of new and novel cytotoxic agents holding promise in HL has led to an “era of riches” for researchers and clinicians working to optimize care for these patients. Among the many challenges facing investigators of HL and other highly curable diseases is the need for clinical trials with large numbers of patients in order to determine the impact of new interventions, as significant differences in efficacy and toxicity may not be apparent in trials with few patients. Without dedicated efforts to coordinate these trials on a broad scale, the true value of these promising new agents may not be realized.
Acknowledgments and Disclosure
Dr. Friedberg is a Scholar in Clinical Research of the Leukemia and Lymphoma Society and is supported, in part, by the University of Rochester SPORE in Lymphoma (CA-130805). He is a member of a Data and Safety Monitoring Board sponsored by Eli Lilly, and previously served on an advisory board for Seattle Genetics. No other potential conflicts of interest relevant to this article were reported.