I’m doing chemo now, Part 2; Choosing the right therapy for my AIHA and CLL

When last we left off in Part 1, it was a Thursday and I was meeting with my hem/onc, Dr. Belle, having ditched the clueless Dr. O’Leary. Hemolysis from my autoimmune hemolytic anemia had been going on for two weeks and huge doses of steroids were only marginally effective. I was feeling weak and anemic, but we had no idea until after my office visit, when blood labs were run at a nearby hospital, that my hemoglobin (HGB) had plummeted to 7.0. What we did know was that my condition demanded immediate attention, and that chemotherapy of some kind would commence the following Monday.

We discussed the options. As readers may recall, I had been undergoing treatment for AIHA since March and had relapsed twice, the most recent event occurring when Dr. O'Leary attempted to step down my steroid to 4 mg every other day. It was becoming clear that my relapses were severe and that I was a candidate for treating the underlying CLL, which is what is done when steroids -- and failing that, Rituxan -- cannot control or resolve the hemolysis, which is the destruction of red blood cells. Let the hemolysis continue unabated and the result is death.

“The general thought on AIHA,” Dr. Terry Hamblin has written, “is that it should be controlled by steroids alone, but if it is impossible to control when the steroids are tapered then the underlying CLL needs to be treated. . . . About 30-40% of AIHA is controlled by steroids alone and does not require any treatment for the CLL.”

Hemolysis in many AIHA patients is more gentle than it has been in me: In the two weeks prior to my arrival at Dr. Belle’s office, my HGB had fallen from 12.4 -- close to normal -- to 7.0, borderline transfusible. Indeed, it reached transfusion territory -- 6.7 -- on Saturday, two days after I met with Dr. Belle.

To FCR or not to FCR, that was the question

The first thought Dr. Belle and I had was to hit it hard, which meant FCR. Despite the risks involved that have been well-documented here and elsewhere, FCR remains pretty much the most thorough and effective treatment regimen available for CLL. Indeed, a CLL expert with whom I had consulted several days earlier had suggested cyclosporin to calm the hemolysis, followed by FCR. However, this suggestion was made before we knew that my HGB was falling so rapidly. And as Dr. Hamblin has pointed out on the ACOR CLL List, “I do recommend cyclosporin, though it takes about six weeks to work.”

That six weeks was the monkey wrench in the plan. I clearly would not last that long, having lost an average of four-tenths of a point of HGB every day for two weeks. Tranfusions would not be a useful bridge here as they are of only temporary use; a unit of blood gives you about a one-point rise in HGB, which in my case would be hemolyzed away in two to three days. I also did not want to extend the stress my heart was under for several more weeks; bottom-of-the-barrel hemoglobin puts you at increased risk for a heart attack.

So, what about jumping right to FCR without the cyclosporin?

Fludarabine (as well as its purine analogue cousins pentostatin and cladribine) is a well-known catalyst for AIHA.This is because, it is believed, the drug suppresses the activity of regulatory T cells, which normally keep killer T cells from running amok and thus contributing to the establishment of autoimmune disorders. There have been cases of rapid and even fatal hemolysis following treatment with single-agent fludarabine. This information, from the 2006 American Society of Hematology Education Book, should raise a big red flag (boldface mine):

“Because of the association of AIHA with chronic lymphocytic leukemia, many people with AIHA can be expected to have been treated with purine nucleoside analogues such as fludarabine . . . In 1995, Myint et al reported on 52 patients with AIHA treated with fludarabine, in whom severe AIHA occurred in 12 (23%) after a median of four courses. Nine of these 12 had no prior evidence of AIHA. Eight were retreated with fludarabine at a later time, and severe AIHA recurred in 6. The authors opined that a disturbance of immunoregulatory T cells was responsible for the problem, in that T-cell lymphopenia is a recognized effect of fludarabine. Weiss et al reviewed this subject, and reported on 24 patients with AIHA following fludarabine therapy for chronic lymphocytic leukemia. Most of the patients developed the AIHA after one to three courses of drug, and seven (29%) died of complications related to the AIHA. Of 8 patients rechallenged with fludarabine at a later time, 7 had recurrent AIHA, and 3 died.”

There you have it again folks, the stark truth: People die of AIHA.

So the question came up: Would it be safe to use fludarabine if it was being infused in the company of cyclophosphamide and Rituxan, both of which have the opposite effect from triggering hemolysis, and both of which are used to combat AIHA?

Well, lo and behold, MD Anderson recently published an abstract on the subject of combination therapy and AIHA, with the pertinent results as follows:

“Fludarabine has been associated with AIHA, whereas both rituximab and cyclophosphamide have been used to treat this condition. Combining these agents with fludarabine may reduce the likelihood of AIHA. We report on the incidence, outcome and pretreatment predictors of IA [immune anemias] in 300 patients treated with fludarabine, cyclophosphamide and rituximab (FCR). Nineteen patients (6.5%) developed IA on or after treatment with FCR. Most patients (82.4%) with AIHA had a negative direct antiglobulin test (DAT). Additional markers of haemolysis (indirect hyperbilirubinaemia, reticulocytosis, low haptoglobin and elevated lactate dehydrogenase levels) confirmed the presence of AIHA in these patients. . . . No haemolytic crisis was observed during FCR therapy in eight patients with AIHA prior to FCR. Thus, the incidence of IA among CLL patients treated with FCR was comparable with historical rates. . . . Pre-existing AIHA need not preclude front-line FCR therapy.”

Good to go, right? Well now, wait a minute. As I have said once or twice, the key in life -- or in battling CLL -- is not to know all the answers but rather to ask the right questions.

Is FCR safe in patients undergoing active, severe hemolysis?

That was the million-dollar question. Look carefully at the MDA abstract. It does not address that concern. It says patients developed AIHA after treatment, and that most of them were Coombs (aka DAT) negative. Eight patients who had experienced AIHA previous to FCR did not have a hemolytic crisis (and do note the use of the word “crisis” here.)

This is all well and good, but I did not fit the description of those patients. I was in the middle of a hemolytic crisis. I was Coombs positive and all those markers mentioned by MD Anderson -- bilirubin, reticulocyte count, haptoglobin, and LDH -- were severely out of whack. I would be going into FCR therapy with active, nasty, life-depleting AIHA and hemoglobin so low that there was little room for error. My life was on the line here and I was determined not to proceed unless I was convinced it was safe. Dr. Belle shared my concern.

Alas, there is precious little information on this subject available on the internet. I have run across anecdotal accounts of hemolyzing patients being treated with FCR and the like. But as my mother used to say, "If everyone else was jumping off a cliff, would you jump off, too?" I wanted information that was a little more solid, and, as anemic as I was, I set out to find it.

I had recently read with interest a paper by Drs. Clive Zent and Wei Ding of the Mayo Clinic in Rochester, MN. Called Diagnosis and Management of Autoimmune Complications of CLL/SLL, it lays out several of the treatment options for autoimmune disorders. This paragraph caught my eye: “Refractory AIHA can be difficult to manage, with no consistently effective treatment options. Splenectomy can decrease hemolysis but is less effective than in management of ITP. An alternative therapeutic approach is chemoimmunotherapy with regimens such as rituximab, cyclophosphamide, vincristine, and prednisone (R-CVP), which combine drugs with single-agent efficacy in the management of AIHA. Despite of lack of published trials, our experience suggests that this therapy is frequently effective.”

Dr. Zent is Mayo’s resident expert on such matters and he was kind enough to correspond with me when I asked for some further information. It should be emphasized here that I did not ask Dr. Zent for medical advice; I was asking him to further illuminate issues raised in his paper and elsewhere. That he did, and it is with his gracious permission that I post his answers here to share with the patient community (any boldface in his comments, or elsewhere, is mine).

“Use of purine analogues (fludarabine, pentostatin or 2CdA [cladribine]) as monotherapy in CLL after AIHA is not a good idea,” he told me.”Use of FCR or PCR is very controversial. Although these combinations have been used without precipitating AIHA in some patients, we have seen recurrence of AIHA and ITP. My preference is to avoid purine analogues whenever possible. Am also concerned about use of alemtuzumab [Campath] monotherapy although we have used alemtuzumab and rituximab with good results in a few patients. Problem at this time is lack of controlled trials of any of these combinations in patients with CLL and autoimmune complications.”

(So, in case you missed it, it is useful to point out that not everyone shares MD Anderson’s confidence that FCR can be administered to patients with a history of AIHA. A patient’s case history, clinical symptoms, and long-term treatment strategy no doubt play a big role in deciding whether FCR is advisable at some point.)

I asked Dr. Zent about R+CD, which is Rituxan, cyclophosphamide, and dexamethasone, a protocol used successfully by Kanti Rai et al to treat steroid-refractory AIHA. It is similar to R+CVP, except that it lacks vincristine. Had they tried R+CD at Mayo, I wondered, and did the vincristine really add that much to it?

“RCD and R-CVP are usually indicated for patients who need treatment for both CLL and their autoimmune complications, less often for "steroid resistant" AIHA in CLL,” he replied. “We have tried the RCD protocol and also use the R-CVP protocol without vincristine for some patients (very similar to RCD). Dexamethasone at 12 mg/dose is equivalent potency and toxicity to the prednisone (40 mg/m2) of the R-CVP protocol.

"Vincristine has very good activity against AIHA and ITP in many patients -- acts against macrophages causing RBC and platelet destruction -- and is thus a useful component of treatment. Vincristine also has activity against CLL. The drug has very little other toxicity and in particular minimal bone marrow toxicity. In contrast both cyclophosphamide and corticosteroids have a wide spectrum of serious toxicities. Obviously patients need to be monitored carefully for vincristine neurotoxicity and the drug stopped early if this occurs. In most cases full recovery is likely.”

(On the subject of vincristine, I followed up with this question: If a patient develops any symptoms of neurotoxicity -- no matter how mild -- is that an indication to stop treatment with the drug? Can the drug be safely used later if the the symptoms resolve, or is it best not to use it again, period?

“There are no data that I am aware of that answer this question," Dr. Zent replied. “My suggestion is that vincristine use can be continued if all neurological symptoms and signs resolve at the time of the next treatment, that the dose should be reduced if minimal symptoms persist and the the drug should be stopped if marked symptoms occur.”)

Finally, I asked this direct question: What do you generally recommend for patients with steroid-resistant AIHA who are actively hemolyzing? Would adding a purine analogue at that point, in the context of FCR or PCR, be acceptable or is it possible to make the hemolysis even worse by doing so?

His reply:

“Would depend on whether the CLL needs treatment on its own merits -- and that is a complex problem. If actively hemolyzing and CLL does not need treatment would consider adding rituximab to corticosteroids and then adding cyclosporin if that does not work. Acute hemolysis treatment options also include IVIG. If actively hemolyzing and needs treatment for CLL as well, would then consider R-CVP, alemtuzumab and rituximab, etc. Using a purine analogue would be very risky even in combination therapy."

"Very risky" was warning enough for me. That answer confirmed my suspicions and led to a new direction in thinking.

Settling on a better choice

So if FCR was out, what would be the next best thing? What would deliver the punch to get the job done without exposing me to risky purine analogues?

There were two candidate therapies: R+CD and R+CVP.

The Rai protocol and its results in a small group of patients is described in a 2002 paper and a follow-up 2006 abstract. All the patients had recovery of hemoglobin, with 50% achieving Coombs negativity. For those patients, the mean duration of response was 23 months, as opposed to 8.8 months for those who remained Coombs positive. “This finding that Coombs conversion portends a longer duration of response suggests treatment goals for AIHA should be a conversion to Coombs negative, and not stopped with recovery of HGB,” the authors wrote. It should be noted here that all the patients in this study were Rituxan-naive, which I am certainly not.

I asked Dr. Zent about the aim of treatment: Do you regard achievement of Coombs negativity as the end goal of treatment? If it is not achieved with one regimen, say R+CD or R+CVP, is it worth following up with another -- say Rituxan plus Campath -- in an attempt to achieve that goal?

“The clinical end point of treatment should be control of hemolysis. This is sometimes achievable with a shorter course of therapy and sometimes needs maintenance therapy (e.g. longer term prednisone). Conversion to a negative DAT (Coombs) test does correlate with a better response. However this is not always achievable and attempting to achieve this goal can cause considerable toxicity in some patients. You may recall that up to 30% of all CLL patients can have a positive DAT at some time during the course of their disease, and only a small fraction will have clinically relevant hemolysis.”

R+CVP is usually used to treat Non-Hodgkins Lymphoma and Follicular Lymphoma and is not a choice, under normal circumstances, for the treatment of CLL. (Indeed, an older study of advanced CLL patients compared chlorambucil + prednisone to CVP and found no advantage for CVP. The addition of Rituxan no doubt increases the effectiveness of both regimens, but there is a reason why R+CVP is not a frontline choice in CLL therapy and why its use is best restricted to unusual cases like mine.)

While not especially marrow toxic and well-tolerated by many patients, vincristine can cause potentially serious problems. These include peripheral neuropathy, central nervous system difficulties, and damage to the optic nerve. (Because of this, some doctors just don’t like it; Dr. Hamblin, in a comment on Part 1 of this post, wrote that he had seen “an old lady confined to a wheelchair for the rest of her life after just one dose.”) However, as Dr. Zent pointed out, vincristine has potent effects against the out-of-control macrophages that chew up precious red blood cells, as well as against the CLL. Clearly, as with any chemotherapy drug, there is an element of risk; the question is, under what circumstances might the rewards outweigh that risk?

Marilyn and I discussed all this, Dr. Belle and I touched base by phone and e-mail, and I also had a long talk with a learned friend. The upshot of all this was to recognize the following: Of the three components of the Rai protocol -- Rituxan, cyclophosphamide, and a steroid -- I had already used two. Rituxan, of which I have had bucketfuls since January 2004, does not work as well in me now as it did early on. While still somewhat effective, it could not be counted on to work as well as it did in those Rituxan-naive patients described in the study. Similarly, I was becoming steroid refractory. The only new element was the cyclophosphamide. Would it be enough, even in synergy with the other two drugs, to stop the hemolysis, help turn my red counts around, and maybe even get me to Coombs negativity?

The chances of success appeared greater by adding another agent, the vincristine. Two new drugs would probably lead to a deeper response than one. R+CVP was the strongest response that I could safely mount given the seriousness of my situation. The lack of a clinical trial was not a big stumbling block as I regard Mayo as one of the very best centers for treating CLL. Their leukemia team is filled with bright, reliable, and upstanding researchers and clinicians. If they feel comfortable recommending it in their article on autoimmune disorders, I feel comfortable trying it.

And so, the choice was made and I began treatment on Monday, October 22. How it has worked out thus far is described in the “How I’m Doing” section below.

One last note: I had long considered that my next step in treating CLL after single-agent Rituxan would be steroids in combination with Rituxan. On a pulsed, occasional basis steroids can help Rituxan by pushing CLL cells out of the marrow and nodes and into the bloodstream, where the monoclonal antibody can more easily get at them. Steroids by themselves can also help relieve painful, swollen nodes, at least temporarily. But, as I have learned this year, long-term, rather high doses of steroids of the sort used to treat my AIHA provide a limited CLL remission and are accompanied by potentially problematic side effects, osteoporosis and loss of muscle mass being the best known.

My next line of defense was to be the addition of an alkalyting agent, which means chlorambucil or cyclophosphamide. The former is not used to treat AIHA and indeed can trigger AIHA as often as fludarabine, though the episodes are not so severe. The latter is used to treat AIHA and can provide an immediate result. That made the choice easy. That Dr. John Byrd had also told me that cyclophosphamide is effective in 11q-deleted patients was a bonus that I hope will net me a decent remission when it comes to the CLL.

Some cautions and conclusions

I have gone into detail here because AIHA is tricky and because many patients and some doctors, such as my now-fired Dr. O’Leary, are at a loss as to what to do about it. The seriousness of the situation is often overlooked. As Dr. Hamblin said in his comments on Part 1 of my post:

“There is no doubt that AIHA is a serious problem in CLL, and if it does not respond to steroids and stay responding then the CLL must be treated. There are no clinical trials that tell us how it should be treated (it would be very difficult to organise one).

“I have personal experience of losing patients to AIHA, but not many doctors have. So you need a doctor who has enough experience to take it seriously. Recently, I have been giving medico legal opinions on young doctors who fail to recognise AIHA in CLL or fail to take it seriously enough.”

My purpose here is to get you, the patient, to take it seriously -- and by extension your doctor. I have provided Dr. Zent’s comments, as well as Dr. Hamblin’s, in the hope that they will help you finesse the issue as they helped me. AIHA is under-discussed on the internet; good information is hard to come by. I hope I have helped to close that gap a little.

And it is a serious matter. Even if I get to Coombs negativity, I cannot count on it lasting forever. Rai’s Coombs-negative patients stayed that way for a median of 23 months. AIHA is now front and center in decisions I make about treating my CLL. It only confirms in my mind the eventual need for a stem cell transplant and in the meantime it dictates the choices I make. FCR, for example, is something I can only risk when Coombs negative and when all other measures, such as haptoglobin, are in the healthy range. Campath by itself is contraindicated for AIHA, which is why experts such as Zent use it in combination with Rituxan. In fact, this warning now appears on the Campath box: “Serious, including fatal, pancytopenia/marrow hypoplasia, autoimmune idiopathic thrombocytopenia, and autoimmune hemolytic anemia can occur in patients receiving Campath.”

In my research I did find a couple of recent abstracts from overseas in which single-agent Campath has been used to treat AIHA; apparently you can play Russian Roulette and win, but it hardly seems worth the risk when adding Rituxan increases the safety margin significantly. (Or as a friend put it, "There are morons everywhere.")

What I found most interesting about these abstracts was the description of the patients, whose AIHA had gotten almost unfathomably bad. I provide this by way of a warning to, again, take AIHA seriously. Here is the start of an abstract from Sweden:

“Progressive B-cell chronic lymphocytic leukemia (B-CLL) is often complicated by autoimmune hemolytic anemia (AIHA), which in some cases may be refractory to conventional therapy such as corticosteroids, rituximab and splenectomy.We report here on 5 patients (median age 66 years, range 59-69) with advanced B-CLL, all of whom developed severe transfusion-dependent AIHA resistant to conventional therapy . . .”

Those people had been through it all and were being kept alive only by blood transfusions. Is that any way to live? Here is another abstract, this time from Hong Kong:

“A 58-year-old man with warm-antibody-mediated autoimmune hemolytic anemia (AIHA) refractory to prednisolone, azathioprine, splenectomy, rituximab and combination chemotherapy, and with unacceptably high transfusion requirement, was treated with alemtuzumab. . . .”

That man could be me if I am not careful, or he could be you.

I asked Dr. Zent the following: Does AIHA ever simply go away for good following treatment, or is a patient who has it once always at risk for relapse?

“The answer to this depends on whether the AIHA was provoked by therapy (e.g. fludarabine) or occurred spontaneously,” he wrote. “Most patients with spontaneous onset AIHA will tend to have a higher risk of relapse -- this is a relapsing condition and needs to be monitored indefinitely. I am sure that a few patients will have long lasting remissions off all treatment, but this will be a rare and retrospective finding.”

In other words, if you’ve come down with AIHA spontaneously -- and there really isn't much rhyme or reason as to who is likely to get it and who isn't -- it is likely to always be lurking in the background.

My final advice is this: If you have pernicious AIHA that gives rise to episodes of serious hemolysis, get yourself to a leading CLL center, such as Mayo, for an evaluation. In my case there wasn’t time and my health insurance doesn’t cover care outside of Arizona, which leads the nation in cacti and skin cancer but not CLL research and treatment. (Within the next year I am going to do something drastic about this.) Do not “hope for the best” and “assume it will all work out OK” because your local doctor says so. (Who knows what would have happened to me had I followed O’Leary’s suggestion that I fight severe hemolysis with just 8 mg of steroids, as described in Part 1.) Do not wait, like some patients I know, until you collapse in the shower or on the street or until your heart fails because it is working well past overtime in an oxygen-starved environment. AIHA can be managed effectively, but it is obvious to me that this is a medical niche, not a common skill, and that you need to find the right people to help.

HOW I’M DOING

So far I have had two cycles of treatment, three weeks apart. The protocol is 375 mg/m2 of Rituxan, followed by 750 mg/m2 of cyclophosphamide and 1.4 mg/m2 of vincristine. As I have also been on methylprednisolone for months, that serves as a substitute for the “P” in the protocol, which calls for 40 mg of prednsione on days 1-5. Since we cannot pulse my steroid, we have kept me on it daily and began to step down once it was clear the hemolysis had stopped. I began the first cycle at 72 mg daily, the second at 40 mg. (Mayo sometimes adds Rituxan on days 8 and 15; given my long history of Rituxan use, Dr. Belle saw little reason to overplay that weak hand.) For the third cycle, which starts this Monday, I will be on 20 mg of methylprednisolone.

The hemolysis came to a screeching halt after the first treatment. Bilirubin and LDH had been high and were normal as of my Nov. 9 blood work. As of Nov. 28, HGB had recovered to 11, hematocrit to 34, and total RBC to 3.09. I am still Coombs positive, though I will do another two cycles at least and can expect a greater likelihood of conversion to negativity as time goes on.

As for the CLL, many lymph nodes have simply disappeared; one node I used to monitor in my neck, which never got much below 3 cm no matter what I did, is now the size of a pea and barely palpable. I can still find some pelvic and abdominal nodes, however; these are reducing but are going to be the last to go, if they go, which I think is increasingly unlikely.

My platelets are at 212, about 50 points higher than they had been, on average, for the last couple of years. The absolute lymphocyte count is way down. It had been at 180,000 before treatment, thanks in large part to the steroids kicking the CLL out of the marrow and nodes and into the bloodstream. It’s now at 20,000 and I can probably expect it to normalize, which is what it did for almost everyone in Rai’s study. Overall, while I am secretly hoping for a CR –- what patient isn’t? -- given the state of the nodes, I will probably get a good PR instead.

There have been downsides. The biggie is that I am highly sensitive to vincristine. Dr. Belle administered only 1.4 mg total during the first infusion, playing it conservatively because of my low HGB and the stress my heart was under. Within two weeks I developed a toxicity that led us to suspend use of the drug.

What I have is the most common side effect, peripheral neuropathy, or numbness in the tips of my fingers. According to a physicians-only website that lists drug side effects, a print-out from which Dr. Belle gave me: “Peripheral neuropathy: Frequently the dose-limiting toxicity of vincristine. Most frequent in patients >40 years of age; occurs usually after an average of 3 weekly doses, but may occur after just one dose.” That’s me. I have a mild case, fortunately, and it should resolve eventually.

I also had, for about two weeks, some mildly blurred vision, which also results from the drug, and which resolved completely.

I wish I had tolerated the vincristine better, since it would do that much more to batter the CLL and send me to Coombs negativity. We may yet use it again in baby doses if the peripheral neuropathy resolves. In the meantime, I am now basically on the Rai protocol, hopefully having gotten some good benefit from the vincristine that first time around. You do what you have to do and hope it works; as I said in an earlier post, sometimes you have to fight fire with fire and either might burn you. I have no regrets and still look forward to normalizing my red counts, perhaps achieving Coombs negativity, and getting a hopefully lengthy break from the AIHA.

I have also been dealing with elevated glucose issues -- the steroids contribute to this, but so does the cyclophosphamide, a delayed side effect of which can be hyperglycemia. I am monitoring my glucose levels daily and have been on a rather Draconian almost-no-sugar, almost-no-carb diet to keep it under control. This, too, should resolve when treatment is completed. (Interestingly, consumption of alcohol keeps the liver from producing glucose, so drinking brut champagne every night with dinner guarantees me a normal result in the morning. What sacrifices we must make for our health!)

For a while we also worried about another side effect of the cyclophosphamide -- somewhat low serum sodium and serum chloride levels, which can lead to kidney problems. At first I was drinking Gatorade Rain to restore electrolyte balance, but it has a lot of sugar and this was discontinued in favor of Inland Sea Water, an electrolyte product Marilyn found that has a high concentration of salt and no glucose, and which resolved these issues in a matter of days.

So, in general, I am feeling much better and managing the side effects of treatment effectively. So far no nausea, which is common with cyclophosphamide, and no hair loss. I don’t have a lot to lose anyway, and I have been losing my hair since I was 13, so alopecia doesn’t freak me out as much as it could.

ABOUT THE ILLUSTRATIONS

The graphics accompanying this post are from a video game called Re-Mission, which is designed for children and teens coping with cancer. It’s a great idea, and a great way to visualize killing cancer cells, which has been shown to be an effective complement to treatment. I imagine adults can play it, too. You can learn all about it here.