Bendamustine: Hype or Hope?

A lot of fuss is being made about bendamustine (also known as Treanda). Because it is an alkylating agent with some of the properties of purine analogues and because it is reputed to work in fludarabine refractory cases there has been a lot of chatter about it. So is it all hype or is it justified?

In August 2007 Cephalon, Inc. announced that the U.S. Food and Drug Administration (FDA) Office of Orphan Products Development granted orphan drug designation for the company's investigational therapy, TREANDA(R) (bendamustine HCl), for the treatment of chronic lymphocytic leukemia (CLL). Orphan drug status is granted by the FDA to promote the development of products that demonstrate promise for the treatment of rare diseases affecting less than 200,000 Americans annually. The orphan drug designation would entitle Cephalon to a seven-year period of marketing exclusivity in the United States for TREANDA, if approved by FDA for the treatment of CLL. Note that the drug is out of patent and there are no development costs to pay for. What Cephalon decide to charge for it should simply compenate them for the costs of bringing it to market. So if it costs more than, say $500 a course, we are being seriously ripped off.

Cephalon’s blurb states that TREANDA is the first rationally designed purine analog / alkylator hybrid, combining the moieties of an antimetabolite and an alkylator. Preclinical data show that TREANDA induces rapid, sustained single- and double-strand DNA damage, which results in apoptosis, or programmed cell death in the tumor. TREANDA also induces mitotic checkpoint inhibition, which results in non-apoptotic cell death. These novel dual-action, anti-tumor effects of TREANDA may be attributed to its unique chemical design.

Cephalon holds exclusive rights to market and develop TREANDA in the United States. TREANDA is licensed from Astellas Deutschland GmbH. Bendamustine is marketed in Germany by Astellas' licensee, MundiPharma International Limited, under the trade name RIBOMUSTIN(R). In Germany, RIBOMUSTIN is indicated as a single-agent or in combination with other anti-cancer agents for indolent NHL, multiple myeloma, and CLL. SymBio Pharmaceuticals Ltd holds exclusive rights to market and sell bendamustine HCl in Asia.

Bendamustine is one of the few drugs to emerge from a Communist country period. It was first synthesized in 1963 in the German Democratic Republic. It is chemically related to the alkylating agent chlorambucil, with the benzene ring in the chlorambucil molecule replaced by a 1-methyl-benzimidazole moiety. The mechanisms of action of bendamustine have been under investigation since the early 1960s, and its first use was as a treatment for multiple myeloma in 1969. It has three active moieties: an alkylating group, in common with the nitrogen mustard family; a benzimidazole ring, which may act as a purine analog; and a butyric acid side-chain – see figures at the top of the article. The drug undergoes extensive first-pass metabolism. However, unmetabolized bendamustine accounts for about 45% of the total drug recovered in urine. The main transformation product is a cytotoxic hydroxy metabolite (beta-hydroxybendamustine). Bendamustine was originally synthesized with the intention of producing an antineoplastic agent with low toxicity and both alkylating and antimetabolic properties. However, it has been shown that, at least at high concentrations, it acts primarily as an alkylating agent.

Bendamustine is certainly a different drug that has activity in a number of tumor types. Our concern must be is it a useful addition to our armory for fighting CLL?
A paper in the Journal of Cancer Research and Clinical Oncology from East Germany in 2001 reported the use of bendamustine in a phase II study of 23 patients with advanced CLL. 13 were chemotherapy naïve, others were called refractory or relapsed, but I don’t have evidence of what that means – and it is certainly critical. The treatment-related mortality was 13%. The overall response rate was 65% and the CR rate was 26%.

A second phase I/II study was published in Haematologica in 2005. This is a rather better journal. The level pf peer review is similar to that of Blood. 16 patients were studied, all had relapsed or refractory disease. Now this needs to be precisely defined. All patients had to be either Binet stage B or C with active disease. They needed to have had at least 1 prior treatment with either chlorambucil or fludarabine – in fact only 2 had had just 1 prior treatment, four had had 2, three 3, four 4, two 5, and one 6. Refractory disease meant relapse within 6 months of previous treatment and they had to have this or relapsed or progressive disease. In fact 13 were resistant to chlorambucil and 4 to fludarabine (there is some overlap but two were apparently not resistant to either).

This was a dose finding study and while they started at 100 mg per square meter twice a month, they had to reduce to 70mg per square meter to find a dose with acceptable toxicity for such patients. The toxicities were mainly hematological. Lymphoma patients can tolerate a higher dose, but they are less likely to have bone marrow impairment.

6 patients were withdrawn from the study because of toxicity. Of the remaining 10, there were 2 CRs, 1 nPR and 4 PRs. The conclusion of this study is that bendamustine is an active agent in CLL and that the safe dose in multiply treated patients is 70 mg/square meter, twice a month. It seems to have activity in patients who do not respond well to chlorambucil, It should be noted that these patients included one patient who had del 11q, but in the majority FISH studies were not available.
I guess the excitement about bendamustine stems from a report at ASH in 2007. A randomized phase III study compared bendamustine with chlorambucil in previously untreated patients. The plain outcome measures appeared impressive: Overall response rate was 69% for bendamustine and 39% for chlorambucil. CR rate was 30% and 2%. Median progression-free survival was 21.7 months and 9.3 months respectively.
But it is important to look at the fine detail in these trials. The dose of bendamustine was 100 mg/sq m on days 1 and 2 and the dose of chlorambucil 0.8 mg/kg on days 1 and 15.

So this is pretty well the maximum dose of bendamustine and a suboptimal dose of chlorambucil. It is not the usual way of reporting chlorambucil doses, but this works out at about 6mg a day for 14 days for an average person, whereas my usual dose is 10mg per day. If we compare the results of the chlorambucil arm in this trial with that of the LRF CLL4 trial that was taking place at the same time. The overall response rate for CLL4 was 72% and the CR rate 7%. The median PFS was 18 months. Now we know that FC is considerably better than chlorambucil, even at the higher dosage, so the real test for bendamustine would be how it compares with FC.

Bendamustine is given as an intravenous infusion, whereas both fludarabine and cyclophosphamide can be taken orally. There are suggestions that bendamustine kills cells by different pathways from those used by other alkylating agents, but as far as I can discover they all require p53, so there is no evidence that it will supplant alemtuzumab in p53 depleted cases.

So at the moment I think it is hype rather than evidence.