Saturday, October 30, 2010

Revlimid and autoimmune hemolytic anemia

When you come down with something like chronic lymphocytic leukemia, you develop empathy for everyone else who has it. And when your CLL triggers autoimmune hemolytic anemia, or AIHA, you become especially empathetic, for you have just joined an elite group.

Depending on whose figures you believe, some 5% to 11% of CLLers will come down with this red-blood-cell-destroying, potentially-life-threatening nonsense. It happened to me early in 2007. (For more details on what it is and how it's treated, read this post. The NIH has a rundown here.) 

I've had a bad time of it, a more severe case than most, becoming refractory to the standard treatments: steroids, rituximab, cyclophosphamide. As of 2009 it was getting so bad that I was relapsing -- in the form of severe bouts of hemolysis of red blood cells -- every few months, no matter what treatment I did. Save for a splenectomy, which is no guarantee of long-term success, I was running out of options.

Along came Revlimid. I have discussed AIHA and Revlimid (lenalidomide) in some recent posts, but the information was buried. 

I don't want to bury the lede here: I am now Coombs negative and have not had a bout of hemolysis since I began Revlimid in March. In my more giddy moments I wonder if I am cured of AIHA.

Revlimid, an immunomodulator drug, has evidently reset my immune system so that it is no longer making antibodies to my red blood cells, a process that leads to hemolysis, which is when macrophages attack those cells. My red blood cell count, along with hemoglobin and hematocrit, has been normal since shortly after I began Revlimid. The Coombs (aka Direct Antiglobulin Test) I had on Oct. 6 confirms that I am negative for those antibodies.

To the best of my knowledge, the effect of Revlimid on CLL-induced AIHA has not been studied nor reported. I believe it is potentially one of the hidden headlines in today's CLL news. Aside from my anecdotal experience, a leading researcher has also found instances where it has been beneficial. In response to an e-mail I sent, Dr. Asher Chanan-Khan wrote:

"I have some experience in context with AIHA. At least 2 patients treated with len while transfusion dependent due to AIHA, resolved their need for transfusion and at least 1 is alive without AIHA 3+ years. Although the numbers are small, it does suggest that AIHA can be controlled with control of disease with len."

What I am saying here to fellow members of the elite club nobody wants to join is this: Consider Revlimid. I was on 5 mg when progress started being made, so it probably won't require a huge dose to control the AIHA. Revlimid may be useful at the very start of your battle as the side effects are usually less problematic than those associated with the traditional first line of defense, steroids. 

There is, of course, no guarantee that Revlimid will work for everyone. There is one report of Revlimid actually inducing warm body AIHA in a woman with diffuse large B cell lymphoma. But that's not CLL.

I also have been on monthly doses of Arzerra during some of this period (ending two months ago). It is possible that the anti-CD20 monoclonal antibody also contributed to my progress, although it appeared to have little effect on the disease and we have now suspended its use. The fact that I was refractory to a similar drug, Rituxan, may explain why Arzerra (ofatumumab) appeared to be providing me with little benefit. 

It would be helpful if some researcher somewhere would undertake a study on Revlimid and CLL-induced AIHA; if the results are what I think they will be, Revlimid could become an important tool in the battle to control CLL, reset the immune system, and perhaps put an end to the scourge of AIHA for large numbers of patients.


Today is January 18, 2011. It has now been almost five months since I quit the Arzerra. Still no sign of AIHA or hemolysis, red counts remain in the normal range, nothing out of whack.

Sunday, October 10, 2010

If pigs could fly, we'd all choose the perfect treatment

There was an interesting exchange in the comments section of Dr. Terry Hamblin’s recent blog post The ultimate FCR advice

The post concerns a pivotal study of 817 patients by the German CLL Study Group. The investigators conclude that there is a survival advantage for most patients who use FCR as frontline therapy and that FCR can change the natural course of chronic lymphocytic leukemia.

The respected study shows that FCR is more effective than FC, meaning that the addition of rituximab is essential to getting the best results. 

The exception to the “FCR is golden” rule is those patients with dysfunctional TP53 pathways, including those with the 17p deletion. FCR needs a functional cell-kill pathway to work, and giving FCR to these high-risk patients is so counterproductive that, Dr. Hamblin indicates, it crosses the line into criminal incompetence.

Of particular interest to my 11q-deleted self is that the our group responded well to FCR. Says Dr. Hamblin: “It is important to recognize that the del 11q patients now fall into this intermediate group. CLL8 has removed indecision as to whether they should be intermediate or high risk. If they are getting FCR they are at intermediate risk.”

All of these are big bullet points in the world of CLL.

After reading Dr. Hamblin’s analysis of the study, one reader was still not convinced that FCR is a good way to go and left this comment:

“Unlike Dr. Hamblin, I don't think FCR should be used as a first-line therapy. Why? Because when the patient relapses (and he will relapse), he will be in a precarious situation indeed. Fludarabine-refractory disease means that the patient has a poor prognosis.

 “My opinion is that there are many treatments now available that are more benign than FCR. These include HDMP+Rituximab, Revlimid+R, flavopiridol, EGCG and others. Clinical trials using such agents as CAL-101, ABT-263, Perixifor+R and others could be considered as well.

“When you use FCR, you shoot the whole wad. Why not try something less toxic to start?”

(I don’t want to get sidetracked here, but except for Rituxan, steroids, and perhaps Revlimid, most of the treatments mentioned above are not feasible for most patients, few of whom have access to clinical trials. EGCG is not a reasonable alternative as it doesn’t appear to do much except in the most indolent of cases.)

Here’s Dr. Hamblin’s reply:

“None of the treatments you mention has ever been compared to FCR so you are asking people to buy a pig in a poke.

“It is true that we don't yet know where to go after FCR, but that is the time for all your experimental treatments.

“I'd rather buy my pig in a poke after I'd been eating roast turkey for 10 years.”

The first thing that ran through my mind is this: It wasn’t that long ago that FCR was the pig in the poke.

* * *

I was diagnosed in 2003, seven years and one month ago. At the time, all I knew is that I was CD38 negative, a good prognostic sign. None of today’s other important prognostic tests -– IgVH mutational status, ZAP-70, FISH -– were available to me. Our CLL researchers were still examining those pigs.

MD Anderson was just starting to report good news about FCR, albeit based on retrospective studies that came in for some criticism. People were intrigued by the results but a little wary of MDA’s pig farming techniques. The German Study confirms that MDA was basically right, even if some of the I’s weren’t dotted and some of the T’s weren’t crossed.

The idea of risk-adapted therapy in CLL was just gaining a foothold. It may be hard for today’s new patients to believe, but most doctors used to make treatment decisions in the dark. They assumed all CLL cases were basically the same,  and pretty much treated everyone with the same thing (chlorambucil, and later fludarabine).  
Clinical symptoms were used to decide when treatment should begin, which is the one thing that has not changed, nor should it. A recent editorial in haematologica includes this comment:

“In spite of this clinically relevant risk hierarchy, the decision to treat is currently not based on the risk profile but on symptomatic disease.  This is important and further supported by the observation that in some subgroups of patients, such as those with 17p deletion (and mutated IGHV), the disease may have an indolent course.”

In other words, we know a lot more thanks to prognostic tests, but the most reliable bottom line resides with symptoms such as swollen nodes, dropping hemoglobin, dropping platelets, frequent infections, and lymphocyte doubling time. Assumptions -- such as “all 11q patients are high risk” or “all 17p patients are screwed” –- may not always be correct. With CLL, the learning curve goes on well into the horizon. The more we know, the more we realize we don’t know.

* * *

So, back in the Dark Ages, I was faced with a decision: FCR or single-agent Rituxan?

My disease looked more like an indolent thing that had finally started getting out of control than it did a goose-stepping disaster. After I was diagnosed, I recalled my last CBC –- from 1996 -– and began to suspect that I had CLL back then. The results showed my lymphocyte count to be a little high, which was chalked up by my doctor at the time to me having an infection, even though I didn’t feel sick.

In 2004, when I got my first FISH result, it gave some credence to the relatively-indolent theory. My result was “normal,” meaning whatever chromosomal damage I had was not on the test. I was not 17p-deleted, nor 11-q deleted.

But because of clinical synptoms –- swollen spleen and nodes and a high lymphocyte count –- the doctors I was seeing seemed to agree that I needed some sort of treatment. Two years later, CLL expert Dr. John Byrd told me I probably could have waited awhile longer, and in retrospect I think he was right. (This is why I suggest all new patients see a CLL expert doctor or two before deciding when and how to treat.)  But at the time I was living in the universe of knowledge that I was living in. 

Do I chose the pig in the poke? Or the other pig in the poke?

FCR struck me as overkill given what I knew about my disease. It was also fraught with questions about side effects and after effects. These weren’t so evident in the early days, and we have learned a lot since: FCR can severely weaken the already weakened immune system, giving rise to a host of problems. Viral reactivation can lead to Richter’s Transformation. Potentially fatal pulmonary problems can develop. Myelosuppression can be severe. Autoimmune hemolytic anemia (AIHA) can result. And so on. FCR has a bit of a Russian Roulette aspect, assuming your gun has about 30 chambers.

The other pig was single-agent Rituxan, also new and virtually untested as a treatment concept. The remission wouldn’t be as good or as long, but the side effects and risks were much more benign. Since it was a soft-glove treatment, it seemed reasonable to start it sooner rather than later, when the disease would be even more out of control.

So that’s the way I went. I responded pretty well for a couple of years, during which I also learned that my disease was worse than I had thought: I tested positive for 11q, I was IgVH unmutated, I was ZAP-70 positive.

* * *

Dr. Byrd also told me that CLL is a long journey, and that the nature of that journey is profoundly influenced by your first treatment.

If I could go back to 2003, armed with the results of the prognostic tests that I have had since, as well as the results of the German Study and other clinical trials conducted since,  would I have chosen FCR?

Probably. Given the full scope of my prognostic markers, the argument in favor certainly would have been stronger. And the weakness that single-agent Rituxan has demonstrated in most patients over the long haul would have made that option less attractive.

Knowing that FCR would be my first treatment, and having had the sense to see Dr. Byrd at the outset, I might have been able to wait another year, perhaps even two, before undertaking it. Judging by the way I first responded to single-agent Rituxan, I probably would have gotten a CR, and the remission probably would have lasted a good three to five years but no longer. We know that unmutated folk like myself relapse faster. That card was dealt when the CLL was born.

Would my FCR experience have been uneventful, giving me years of “worry-free” living? Or would I have not dodged all the bullets that come with firing that big gun?

I have enjoyed relatively good quality of life along the path that I did choose, with the exception of coming down with AIHA in 2007. Might FCR have spared me that? Or might it have caused it to happen, as it recently did to a fellow patient I know?

Might the 11q have been avoided? Or might 17p have resulted from FCR’s selection of CLL clones with dysfunctional TP53 pathways?

Assuming all had gone well, I’d likely be in relapse now, on my second treatment, perhaps even a third. Ironically, I might be on Revlimid to keep the disease in check, which is exactly where I am.

Either path would have led me to the same place, namely the likelihood of needing a stem cell transplant to survive in the long term.

Either path is likely to have prolonged my life. The German Study cannot answer long-term survival questions: How many FCR patients will be alive after, say, 10 years, compared to others? A study in the 1990s showed that single-agent fludarabine led to longer remissions than chlorambucil, but as time went on, it became clear that fludarabine did not provide longer overall survival. Rituxan will probably make a difference in the statistics. But will it have to be part of a chemoimmunotherapy regimen (FCR) to make that difference, or will using it as a single agent also have an impact? 

I said before that I'd probably choose FCR if I could go back in time. It would be the logical thing to do given my prognostics and the options available in 2003. (Would I have tried  Revlimid first had it been a choice? Hmmm.) But there is no guarantee that, had I chosen FCR, I would be in any better shape than I am today.

My guess is that for patients with “mild” cases of CLL, FCR will prove more of a game changer than it will for us “intermediate risk” folk. It makes sense that the more you beat down something that doesn’t grow very fast, the longer it will take to come back.

Will a study conclude in, say 2018, that FCR increases survival of 11q-deleted, unmutated patients by three years, or three months, or not at all? (For that matter, will 11q deletion be an antiquated measure, having given way to some new, more specific way to understand what is driving an individual’s CLL?)

My point is that, as my experience demonstrates, CLL choices are ever-shifting. Nothing is frozen in time. And treatment results can vary greatly by individual because, on some level, CLL cases are as unique as fingerprints.

We are at the beginning of the era of risk-assessment, not the end of it. For all our knowledge -– and the light shed by the German Study should not be discounted -- we are still playing a guessing game. Given everything we don’t know, everything is still, to one degree or another, a pig in a poke.

So judge your pigs as best you can and hope that the one you choose doesn’t turn out to be a turkey.