Bendamustine Toxicity in FL Raises Eyebrows -- and Questions

Alexander M. Castellino, PhD

December 19, 2016

Bendamustine (Treanda, Teva Oncology) is a relative newcomer in the treatment of lymphoma, becoming a popular option for use in combination with rituximab (Rituxan, Genentech/Roche) and replacing more complex three- or four-drug chemotherapy regimens after clinical trial results showed it to be both more effective and less toxic.

It was therefore a big surprise to lymphoma experts to hear new data ― presented during the plenary session at the recent American Society of Hematology (ASH) 2016 Annual Meeting ― showing more deaths with induction regimens containing bendamustine than in other induction regimens that contained combination chemotherapies (eg, CHOP [cyclophosphamide/doxorubicin/vincristine/prednisone], CVP [cyclophosphamide/vincristine/prednisone]).

Indeed, one lymphoma expert described the new data as a bit of a "bombshell" — raising more than a few eyebrows and many questions about a regimen the community has become accustomed to using.

Given that bendamustine with rituximab (R-bendamustine) is a valid treatment choice in newly diagnosed follicular lymphoma (FL), Medscape Medical News asked several lymphoma experts for reactions to the new findings and how these might affect the management of patients with advanced FL.

New Data Show Higher Mortality

The higher bendamustine-associated mortality was reported during the presentation of the GALLIUM study, as reported by Medscape Medical News.

This was a phase 3 head-to-head comparison of the newer CD-20 antibody obinutuzumab (Gazyva, Genentech/Roche) vs the older agent rituximab. The strategy was to use induction therapy with a CD-20 antibody (obinutuzumab or rituximab ) and chemotherapy (bendamustine, CHOP, or CVP) for 6 months followed by maintenance therapy with the CD-20 antibody for 2 years.

Results reported at the ASH meeting showed that patients in the obinutuzumab arm had significantly improved progression-free survival (PFS) when compared to those in the rituximab arm, but there was no difference in overall survival, and there was more toxicity in the obinutuzumab arm.

A further analysis of the toxicity data showed an excess of deaths in the arms that had used bendamustine for induction when compared with the other chemotherapy regimens. Bendamustine-induction regimens were associated with the highest rates of death: 5.6% (19 patients) for obinutuzumab-bendamustine (G-bendamustine) and 4.4% (15 patients) for R-bendamustine. In contrast, deaths occurred in nine patients across all other arms.

Lead investigator Robert E. Marcus, MD, of King's College Hospital, London, United Kingdom, told Medscape Medical News that the fatal toxicities were mainly infections, but that there were also some cardiac, neurologic, and respiratory toxicities that were fatal.

Dr Robert Marcus

The rate of death from infection was also higher for bendamustine-containing regimens — 2.7% (nine patients) for G-bendamustine and 0.6% (two patients) for R-bendamustine. Death from infection was reported in one patient across all other arms.

The new findings give some pause for thought, as they show a greater mortality with R-bendamustine and G-bendamustine induction regimens when compared with R-CHOP and G-CHOP or R-CVP and G-CVP.

Bendamustine Toxicity Was Unexpected and a Surprise

Discussing the safety information, Dr Marcus told Medscape Medical News: "We did not know what to expect."

Dr Marcus explained that after its approval, bendamustine was adopted quite rapidly in North America and Europe on the basis of the BRIGHT and German StiL NHL-1 2003 trials. "There were no long-term safety data, and there were certainly no data on bendamustine followed by rituximab maintenance. But bendamustine was better than CHOP with R induction, and that's why many centers participating in the GALLIUM study chose to use this regimen," he said.

These data from the GALLIUM study "are the only data for bendamustine with antibody followed by maintenance. The results were unusual. Once we saw the mortality data, we wanted to analyze them more deeply. We determined that a subset analysis was warranted, and found that G-bendamustine was no more toxic than R-bendamustine, but they were more toxic than CHOP or CVP-based regimens," he said.

New Data Raise Questions

Nadia Khan, MD, assistant professor in the Department of Hematology-Oncology at the Fox Chase Cancer Center and Hodgkin Lymphoma Panel member, the National Comprehensive Cancer Network, provided initial comments on the GALLIUM study in the previous report by Medscape Medical News.

Dr Nadia Khan

Continuing that discussion, she said that the bendamustine toxicity reported at ASH 2016 took attendees by surprise. "The results were very surprising because we have a lot of prior experience with bendamustine induction and have found it to be safe. The assumption was that because bendamustine-containing regimens were less toxic as compared with R-CHOP, the combination of induction and maintenance would be similarly well tolerated," she said.

"We were taken aback with the bendamustine toxicity reported in the G-bendamustine and R-bendamustine arms compared with the other arms in the study," she added.

The safety data from GALLIUM raise more questions about maintenance after bendamustine induction, she said.

"I was quite surprised by the toxicity [reported for the bendamustine arms in the GALLIUM study]. It has not been my experience," Bruce D. Cheson, MD, professor of medicine, head of hematology and deputy chief of hematology-oncology at the Georgetown Lombardi Comprehensive Cancer Center, Georgetown University Hospital, Washington, DC, told Medscape Medical News.

Dr Cheson uses bendamustine for most patients with previously untreated FL. Although the data from the GALLIUM study are compelling and support use of obinutuzumab because of the advantage in PFS, its efficacy is counterbalanced by a higher level of toxicity, he noted.

"There may be a select patient population, such as the younger, fit patient, in whom I may use G-bendamustine in preference to R-bendamustine," Dr Cheson said. "However, for most, R-bendamustine remains my standard," he said.

For patients in whom there is a suspicion of occult transformation, Dr Cheson uses R-CHOP as induction in preference to R-bendamustine.

"The GALLIUM study provided a lot of information. However, the combination of bendamustine and rituximab is still a standard approach in clinical practice that has been backed by results from other studies," Stephen Ansell, MD, PhD, chair of the Mayo Clinic Lymphoma Group, Rochester, Minnesota, told Medscape Medical News.

Dr Stephen Ansell

"Data with obinutuzumab have not been very consistent in that it has not always proven better than rituximab," he added.

Dr Ansell treats newly diagnosed FL patients either with CD-20 antibody (rituximab) if they have low-risk disease and low-grade symptoms, or with immunochemotherapy (R-chemotherapy) if they are high-risk symptomatic patients. He does not typically use maintenance therapy in low-risk patients.

"Induction followed by maintenance is usually reserved for high-risk, bulky disease for relapsed/refractory FL," Dr Ansell said.

"Bendamustine is associated with infusion-site reactions, and although it has some toxicity, it is not difficult to give," he noted. However, because obinutuzumab seems to have more toxicity than rituximab, G-bendamustine appears more toxic than R-bendamustine, and exporting G-bendamustine into clinical practice may be more challenging, Dr Ansell explained.

"A barrier to obinutuzumab uptake could be the side effect profile. Giving G-bendamustine is a legitimate concern, and I am not entirely convinced that the G-regimen is a better regimen," Dr Ansell said.

Anas Younes, MD, medical oncologist, chief, Lymphoma Service, Memorial Sloan Kettering Cancer Center, New York City, indicated that the results of GALLIUM would not change his clinical practice. "Patients with newly diagnosed FL are currently treated with R-based chemotherapy. R-bendamustine is one of the most widely used regimens," he said. "However, there are no data on the safety and efficacy of R-bendamustine followed by rituximab maintenance in these patients," he added.

Dr Anas Younes

"While the safety and efficacy of G-bendamustine followed by G maintenance was established in relapsed FL, the safety and efficacy of this regimen in the frontline setting was not previously reported," Dr Younes told Medscape Medical News.

"The GALLIUM study reported that patients who were treated with bendamustine-based regimens had the highest toxicity, including treatment-related death," he iterated. "Bendamustine is an immunosuppressive agent, and it is not surprising to see more side effects with bendamustine-based induction regimens," he said.

For newly diagnosed patients with cardiac risks and relatively low ejection fraction, R-bendamustine and G-bendamustine are both valid options, he continued. For these patients, maintenance therapy should be avoided to reduce toxicity, Dr Younes explained. "For patients with no cardiac risk and who are candidates for maintenance therapy, it would be safer to consider R-CHOP or G-CHOP regimens," he said.

Dr Khan agreed. "However, if you use maintenance and would like to continue maintenance after induction therapy with a bendamustine-containing regimen, results of the GALLIUM should give one a cause to ponder one's strategy," she said.

"Not all patients are candidates for maintenance therapy," Dr Khan told Medscape Medical News. On the basis of data from the PRIMA study, Dr Khan reserves maintenance for patients who do not achieve a complete response with induction therapy.

PRIMA compared the strategy of rituximab maintenance vs no maintenance after induction with R-chemotherapy. When PRIMA was initiated, bendamustine had yet to be approved, and CHOP and CVP were chemotherapies of choice in PRIMA.

"PRIMA indicated that there is a PFS advantage with maintenance, and GALLIUM indicated that obinutuzumab and rituximab have equivalent efficacies through the 6-month induction phase," Dr Khan said.

Noting that the Kaplan-Meier curves separated during the maintenance phase, she explained that she might continue treating patients with rituximab-based induction and, in select cases, provide obinutuzumab as maintenance therapy.

"However, due to the excess toxicity in bendamustine-containing induction regimens, the issue I am likely to face in my practice is the ability to identify patients with aggressive disease who are less likely to achieve a complete response with induction therapy," she said. "Since I may provide maintenance to these patients, I need to identify these patients early so that I can provide them CHOP-based induction regimens," she explained.

"The data from GALLIUM will not change the way I practice induction therapy," Dr Khan said. "However, I have no data to guide me for immunotherapy maintenance after bendamustine-based induction," she added.

Because maintenance immunotherapy does not improve overall survival, one has to rethink one's strategy after bendamustine-based induction for patients with more aggressive disease, Dr Khan pointed out. "If there is no survival benefit with maintenance after chemotherapy, how does one justify a strategy that may even be slightly more harmful, such as with bendamustine-based induction followed by maintenance?" Dr Khan asked.

...And Dr Marcus Responds

"Even if induction followed by maintenance is not associated with an overall survival benefit, achieving the longest survival with fewest relapses is what I strive for," Dr Marcus told Medscape Medical News.

"Remission is good; relapse is bad. The longer a patient is in remission, the better it is," he added.

He had additional thoughts for practicing physicians to ponder. With respect to the choice of chemotherapy in induction, he emphasized that it depends on individual practice settings. "If you've seen the same side effects we saw in the trial, then be cautious about that chemotherapy," he said.

Dr Marcus, however, believes in chemoimmunotherapy induction followed by immunotherapy maintenance. In his own practice, Dr Marcus uses R-bendamustine induction followed by rituximab maintenance, which does not seem to be associated with major toxicity in his patient population. "Now we will have to be more careful," he added.

Obinutuzumab is not currently approved for use as first-line treatment of lymphoma, only for use in relapsed and refractory cases. If it does receive approval for front-line use, Dr Marcus says he would be in favor of switching to G-chemotherapy induction followed by obinutuzumab maintenance, perhaps with the nontoxic CVP regimen.

"The GALLIUM trial was stopped by the data-monitoring safety board because it was viewed as positive, and the trialists also viewed it as such," he said.

He also noted that G-chemotherapy, regardless of the chemotherapy backbone, was associated with the highest likelihood of achieving minimal residual disease negativity — results also reported at ASH. Indeed, all three chemoimmunotherapy-induction regimens provided similar minimal residual disease (MRD) response rates (~95%). Achieving MRD negativity with R-chemotherapy was dependent on the chemotherapy backbone, he explained.

"Additionally, achieving MRD negativity at the end of induction was associated with longer progression-free survival," he said.

Dr Khan offered one final thought that clinicians have to mull over with respect to future trials. "The higher mortality in the bendamustine-based induction arms is going to require more understanding before developing more trials around this backbone," she noted.

Dr Marcus consults for Roche and receives honoraria from Takeda. Dr Ansell receives funding from BMS, Seattle Genetics, Merck, Celldex, and Affimed. Dr Cheson is on the board of directors or advisory committees of Pharmacyclics and Acerta and receives research funding from Pharmacyclics, Acerta, and Gilead. He is also on the advisory board of Medscape Oncology. Dr Khan is on the advisory board of AbbVie and Pharmacyclics.

American Society of Hematology (ASH) 2016 Annual Meeting. Abstract 6, presented December 4, 2016.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.