Friday, December 17, 2010

Revlimid dosing

One thing I'm learning about Revlimid and CLL is that it's a bit like the Wild West. There aren't a lot of rules, you can strike it rich or get gunned down, and you'd better be a little tough to survive what life throws at you. A shot of whisky now and then doesn't hurt, either.

Deciding on a dose (of Revlimid, that is) and how frequently to take it is a particularly open-ended subject. In one clinical trial, the goal is to get patients up to 25 mg per day, if they can tolerate it. Another study suggests that 2.5 mg to 5 mg may be the maximum tolerated dose for heavily pretreated patients. Based on reported data, Celgene Corporation, the maker of the drug, advises that more than 10 mg is potentially unsafe for CLLers. And now there's an abstract out about pulsed dosing in relapsed patients that suggests 20 mg is more effective than lower doses.

The study was done at the NIH and an abstract was presented at the recent American Society of Hematology meeting. Patients were given Revlimid three weeks on and three weeks off. Thirty-one patients had at least two cycles of therapy, which is a short period for Revlimid. Some had as many as eight, but the abstract does not address the median number, which is significant in that Revlimid can require long exposure for the best effects. 

At any rate, there were no CRs; 16% had PRs, 58% stable disease, and 26% progressive disease. Of the five patients who achieved a PR, four had deletion 17p and bulky disease. Four of that five started Revlimid at 20 mg; one started at 10 mg. Pulsed dosing did not lead to fewer toxicities. Once treatment was stopped, median time to next therapy was  six months, although it ranged from two to 18 and was significantly better in the PR group. The full abstract is well worth reading and can be found at the bottom of this post as well as, hopefully, through the link.

There is no "written in stone" rule about Revlimid dosing, but something of a consensus may be emerging. Based on the NIH study and other data, not all of it published, we do seem to be learning that the highest tolerable dose (up to 25 mg) may be most effective. But getting there can be a rocky road, pardner, which I will blog about within the next few weeks. Celgene is not wrong to suggest that the higher you go, the more trouble you may get into. 

And staying there, or staying at any dose level, is not guaranteed. Side effects -- low neutrophils, low platelets, serious rash, blood clots, etc., etc. -- can derail the Revlimid train.  As the NIH study points out, Grade 3 or 4 neutropenia was seen in 56% of cycles, "often worsening in continuing cycles." One thing we do know for sure about Revlimid is that individual response can be unpredictable, both in terms of effectiveness and side effects. 

Revlimid is effective in many patients who have become refractory to other drugs, but can you become refractory to Revlimid? Apparently, yes, from what I've heard, although I don't think there have been any studies on the subject. This has happened to some patients, mostly those who have been heavily pretreated. Why this happens to some and not others, and whether it has something to do with dosages and treatment schedules, is unclear.

For those of us who achieve a remission or stable disease with the drug, how do we maintain it? Again, there's nothing set in stone. In one leading center they keep you on the highest dose you can tolerate for a year, then take you off. You're monitored and only resume Revlimid again when you begin to relapse. There aren't reports of patients becoming refractory there, so perhaps this method has a hand in maintaining the drug's usefulness.

One fascinating question is whether the Revlimid can actually train the immune system to attack CLL cells on its own, without the Revlimid. This possibility has been suggested by some serious people, but it will probably be a long time before we see any hard data.

Basically, there's still a whole lot of guessing going on when it comes to how much to take and how often. The advantage of taking more, namely better response, can be offset by worsening side effects.  And, of course, one patient can do well on 5 mg when another really needs 15 or 20 to show progress.

Revlimid, like exploring unknown territory, requires that your eyes and ears be open. Back in the old days, not all maps were drawn, and not all were accurate. If you're on Revlimid, you're a trailblazer, like it or not. 

Phase II Trial of Pulse Dosed Lenalidomide In Previously Treated Chronic Lymphocytic Leukemia

Georg Aue, M.D.1, Susan Soto, RN2*, Janet Valdez, PA1*, Diane C Arthur, M.D.3, Xin Tian4* and Adrian Wiestner, M.D., Ph.D.5

1Hematology Branch, National Heart, Lung, Blood Institute,, National Institutes of Health, Bethesda, MD
2National Institutes of Health, Hematology Branch, National Heart, Lung, and Blood Institute, Bethesda, MD
3Laboratory of Pathology, NIH/NCI, Bethesda, MD
4Biostatistics, National Heart Lung and Blood Institute, Bethesda, MD
5Hematology Branch, National, Heart, Lung, and Blood Institute, Bethesda, MD

Introduction: Lenalidomide (L) has activity in relapsed chronic lymphocytic leukemia (CLL). The mechanism of action is not well understood but may involve stimulation of anti-leukemic immune responses. Myelosuppression especially neutropenia is a concerning side effect. We reasoned that pulsed dosing of lenalidomide could reduce myelosuppression while maintaining the immune stimulatory effect. To test this concept we initiated a single center, phase II trial ( Identifier: NCT00465127) of lenalidomide given in cycles of 3 weeks on, 3 weeks off drug (42 day cycles).

Methods: Patients (pts) with relapsed CLL or small lymphocytic lymphoma with ANC>500/ul and platelets >20,000/ul were eligible. The primary endpoint defined as response after 4 cycles has been recorded for all participants. Pts with partial response were allowed to receive up to 4 additional cycles. The starting dose for the first 10 pts was 20 mg daily; the starting dose for pt 11 onwards was lowered to 10 mg daily because of toxicities observed in other L trials for CLL. TLS prophylaxis with Allopurinol was mandated during cycle 1-3. Deep venous thrombosis (DVT) prophylaxis was not mandated unless risk factors were present. Ibuprofen and corticosteroids were allowed to treat symptoms of a cytokine release syndrome (CRS, defined by LN swelling, fever, fatigue, pain, chills, dehydration). Responses were assessed by IWCLL criteria and included CT scanning.Patient characteristics (n=33) were: median age 64 years (36-78); median number of prior therapies 3 (range 1-7); 52% Rai stage III-IV; 70% bulky disease; 30% fludarabine refractory; 56% (of 27 pts) ZAP70 pos; 64% (of 25 pts) unmutated immune globulin VH mutation status; 43% del 17p; 15% del 11q.

Results: A total of 131 cycles of L were given. 31 pts received at least 2 cycles of therapy (range 2-8) and were evaluable for response: 5 (16%) partial response (PR), 18 (58%) stable disease, and 8 (26%) progressive disease. 4 of 5 responding pts had del 17p and bulky disease. In responders (n=5, PR) vs non responders (n=26, SD+PD) the PFS was 16 vs 6 months (p>0.01), and the time to next therapy was 17 vs 6 months (p>0.01), respectively. Once treatment was stopped, duration of response was short lived (median 6 months, range 2-18). 4 out of 5 responders were observed in the 20 mg dose starting group versus only 1 responder in the 10 mg group (p=0.03). There was no difference in the CRS score between the 2 groups (2.5 vs 1.5, p=0.17). Hematologic responses were observed in 11 out of 24 CLL pts (45%). At the completion of 4 cycles CD4 and CD8 counts increased by 20%, while NK cell counts remained unchanged. Dose modifications/withdrawl: 41% of cycles required dose adjustments prior to or during cycles 1-4. 9 pts (27%) did not complete 4 cycles of L because of: autoimmune cytopenias (2 pts), side effects (4 pts; CRS 1 pt, neutropenia 3 pts), withdrawal from study (2 pts), and disease progression (1 pt). Toxicity: Gr 3/4 neutropenia was observed in 56% of 131 cycles, often worsening with cumulative cycles. Gr 3/4 thombocytopenia and anemia were seen in 30% and 15% of cycles, respectively. Gr 1/2 and 3/4 infections occurred in 23% and 11% of cycles, respectively, 8 of those in the setting of neutropenia. Gr 3 CMV colitis, PCP pneumonia and Candedemia each were observed once. 1 patient died from streptococcal sepsis in cycle 4. Gr 1/2 and 3/4 CRS were observed in 43% and 10% of cycles, respectively. A CRS was encountered in 78% of first cycles typically within the first week, and in 48%, 38% and 30% of cycles 2-4, respectively. 6 DVTs (Gr 3) were diagnosed in 5 pts. Other common side effects were fatigue (62%), rash (39%) and muscle cramps (27%), all Gr 1/2. No case of tumor lysis syndrome was seen.

Conclusion: L cycled 3 weeks on, 3 weeks off led to stable disease in the majority of pts and induced PRs in 16% of relapsed CLL patients with high risk disease. Pulse dosing of L did not lead to reduced toxicities. Myelosuppression and infections remain a major concern. 4 out of 5 responders were observed in the 20 mg cohort arguing for higher L starting doses. Notably, side effects, particularly the CRS, were similar in the two cohorts. Once L was discontinued, the duration of response was short, suggesting a need for continued therapy in pts who are able to tolerate the drug.

Disclosures: Off Label Use: Lenalidomide is not FDA approved in Chronic Lymphocytic leukemia.


ozark Wayne said...

Hi Dave.
In your report of this trial, is CRS another name for 'tumor flare?'

Anonymous said...

Hello David...Sometimes all that we have are poor choices. I certainly understand why you began using lenalidomide and why you have continued to take it. Your experience mirros the data that is reflected in study after study...lenalidomide, while helpful to a certain extent, overall is a lousey drug for use in CLL/SLL.

In your case, stable disease and absence of recurrence of AIHA have been positive results despite your side effects.

I note that 2 patients in the study cited dropped out because of immune cytopenias, so it seems likely that lenalidomide is not a slam dunk means to avoid these problems or their recurrence. The side effect profile of the patients in the study cited seems substantial to me.

As with all therapies, there is always a price to pay. My sense is that the price overall with lenalidomide is steep and doubt that lenalidomide will be used much for CLL 5 to 10 years from now.

I hope that you continue to benefit from it for the forseeable future.

~chris said...

Low dose lenalidomide may find a niche as a immune system booster and perhaps as a treatment for autoimmunity...

David Arenson said...

CRS involves a number of things, as the abstract mentions, but tumor flare ("LN swelling") is indeed probably the most serious of them.

The second commenter has some good points, although I don't agree completely.

One thing I see over and over is that the effectiveness of Revlimid and the severity of side effects varies greatly from individual to individual. So not everyone pays the same price for this therapy.

It's also important to remember that Revlimid is kinder than fludarabine and the like when it comes to some significant concerns: immunosuppression, mutagenic issues, and possibly becoming refractory over the long term. As such it can be argued that while it provides less deep remissions, it can keep the disease in check without much of the damage associated with regular chemo. This means it may not be so lousy for CLL after all.

Revlimid also works as well on high-risk (17p) CLLers as on others. This is significant for those patients, who have limited treatment options.

Revlimid is proving itself useful for some relapsed, refractory patients. Since there is little that works on these patients after they have failed chemo, that is also significant.

Since it is often infections that kill CLLers, especially after their immune system has been pounded to nothing by both the disease and the cure (chemo), that fact that Revlimid helps improve the immune system may also be significant. (This dovetails into Chris's point.)

It is hard to say what Revlimid's role in CLL will be in 5 or 10 years. It all depends on whether better things -- more effective therapies with less toxicity -- come along.

In the interim I think it serves some useful purposes for those on whom it works.

A final but interesting question: As a frontline treatment, can it keep the disease in check while preserving and enhancing the immune system? And will this allow these patients longer life without significantly jeopardizing their ability to respond to traditional chemo should that be needed down the road?

Anonymous said...

It's the "donut vs hole" debate. You have seem the downsides of L as you will soon blog on again. I suspect that you've been holding off until you get back to a reasonable dose. The fact that you haven't been able to tolerate a larger dose goes along with my sense that L is a louder therapy. Perhaps better than some, but louder nevertheless. If a person is going to take any treatment, let alone one as expensive as L, at least they should expect a high likelihood of a great response or a low likelihood of serious side effects.

Anonymous said...

I am a 66 yr old risk patient, that has been participating in the Rosewell L study for the last 10 months. I could not tolerate 20 or 25mg due to CIDP & exhaustion but have done very well with 15 mg for the last 5 months. My whites have been within range for 4 months & platlets stayed within range, I have slighlty swollen nodes & my RBC (2ish), HCT & HG (7ish) are low enough to require a transfusion. I had the rash, elnarged nodes & pains in the begining but that all disappeared within the first month. Other than shortness of breath (last 3 weeks) due to low RBC, I feel fine. I will be ending the drug portion of the study in early March, which is when the rubber will hit the road.