Date: January 26, 2005
by Professor Terry J. Hamblin
Royal Bournemouth Hospital
Editor's Note: Professor Terry Hamblin needs no introduction to CLL patients. His long career has focused on asking the key questions in CLL and structuring clinical trials to answer some of them. His seminal work on the prognostic value of the IgVH gene mutation status has become a cornerstone in the risk stratification of CLL patients. We are delighted that Professor Hamblin accepted our invitation to write this article specifically for readers of CLL Topics.
In this article Dr. Hamblin discusses criteria for initiating treatment as well as what treatment can and cannot do for the patient. We heartily second his call for Randomized Controlled Trials (RCTs) to answer many important questions that have been left unanswered by phase II trials. There is so much diversity among CLL patients that anecdotal information or statistically under-powered studies with inadequate controls will only slow down the development of real therapeutic options with well understood risk/reward profiles. Dr. Hamblin makes a strong case for randomized controlled trials. To that we would add a plea for prompt publication in peer-reviewed journals of detailed results from the phase II chemo-immunotherapy combination trials. Prompt publication and full disclosure are the only acceptable norms for clinical trials involving human subjects, in our opinion.
We are grateful to Professor Hamblin for this assessment of where we are in the campaign to cure CLL. Candid assessments such as this are important. How else will we get a fix on where we are, how far we have come and how much further we need to go?
Chronic lymphocytic leukemia is deceptively easy to treat and very difficult to cure. The good news is that many patients, perhaps the majority, will not have their life shortened by having CLL.
It can behave like other types of leukemia in filling up the bone marrow, and diverting production from normal blood cells into the leukemia cells. This results in anemia, low platelets and low neutrophils, opening up the possibility of bleeding or infection. This used to be a common outcome, but nowadays the vast majority of patients are diagnosed at an early stage and only a minority progress to the advanced stage that leads to bone marrow failure. Sometimes patients become anemic with low platelets because the normal blood cells get trapped in an enlarging spleen. This is not the same as bone marrow failure, but it may still be a reason for treatment.
It can transform into an aggressive form of lymphoma. This is called Richter’s syndrome. These aggressive lymphomas get into awkward places like the brain and bones and do a lot of damage. This is quite different to what happens with CLL lymph nodes. These may grow quite large; they may cause symptoms and be unsightly; but they don’t usually erode vital organs or cause major obstruction anywhere.
Enlarged lymph nodes can make you feel unwell. They often produce fevers and sweats. They can make you lose weight and they can cause difficult to explain symptoms like itching and cramps. The cause of these symptoms is the secretion of cytokines like interleukin-1 and Tumor Necrosis Factor. Sometimes these are made by the CLL cells themselves and sometimes by the body’s reaction against them. Symptoms like this are commoner if the CLL has transformed to Richter’s syndrome. Historically, Richter’s syndrome used to occur in 2-3% of patients, but it seems to be getting commoner, with recent reports claiming 8%.
The other way it kills you is by interfering with the immune system. It does this either by making the immune system overactive or underactive. If it becomes overactive it starts attacking your own body rather than just infecting germs, so-called autoimmunity. About 15% of patients will develop autoimmune disease during the course of their illness, almost all of them hemolytic anemia. The next commonest type is immune thrombocytopenia where the platelets are destroyed (about a third of these patients also have hemolytic anemia). All other forms of autoimmune disease are vanishingly rare. Autoimmune disease can usually be successfully treated.
An underactive immune system is almost universal in CLL. In many patients this can be demonstrated by measuring the serum immunoglobulins (this is the reservoir of antibodies produced in the past), but even in patients with normal immunoglobulins the immune system does not function well. Vaccination against common germs like pneumonia and influenza is usually not very successful with poor responses in the majority. Late in the disease it is common for patients to succumb to infection.
What CLL does not do is kill patients because the white cell count is too high. Although with other forms of leukemia a very high white count can cause a stroke because the blood cells sludge in the blood vessels of the brain, this does not happen in CLL even when the white count is over a million. Nor does it kill because of large lymph nodes. Unless Richter’s transformation has occurred, the glands are soft and mobile and don’t seem to press on anything vital.
There are two reasons for treating patients with CLL: 1) to make people live longer 2) to relieve symptoms. Of course it is possible that these two aims may have mutually exclusive results. It may be that patients could live longer yet suffer intolerable symptoms as a result. Alternatively, relieving symptoms might shorten life.
We are much better able to pick those patients who will definitely need treatment in the future from those who will be most unlikely to need any. We can do this using prognostic markers – VH gene mutations, CD38, ZAP-70 and FISH. These tests don’t all give the same result, so ideally all four should be used. It is important to realize that only VH mutations stay the same throughout the course of the disease. CD38 and FISH certainly can change for the worse. It’s too early to say whether ZAP-70 might change.
The National Cancer Institute has issued guidelines on who should be treated. These suggest that one of the items on the following list must be present:
It has to be said that many patients are treated for reasons that fall short of these guidelines.
We have known for some time that patients with early stage CLL do not benefit from treatment with chlorambucil before they develop symptoms. Whether this is true for treatment with fludarabine or combinations of drugs including fludarabine has never been tested, so we do not know.
It would be reasonable to conduct trials to answer this question.
So far there are no treatments that regularly cure patients with CLL. There are certainly patients who have been in remission for a very long time who might be cured, but since we know that some patients have very slowly progressive leukemias we cannot predict the future.
Stem cell transplants are believed to hold the most promise. Autografts, where the stem cells are harvested from the patient himself or herself after chemotherapy has removed most or all of the CLL from the bone marrow have been studied carefully by the German group. They have shown that patients with unmutated VH genes are not cured by an autograft. By four years all patients had detectable disease in their bone marrow. For patients with mutated VH genes the outlook was better, but we know that this is a much slower growing version of CLL and follow-up may not have been long enough to detect relapse. Furthermore, these patients’ disease might well have been not severe enough to justify such extreme treatment. The British autograft trial is a bit more alarming because they have reported a worrying incidence of myelodysplastic syndrome (MDS) among those treated. MDS is a bone marrow disease where the blood cells don’t work properly and in about a third of cases MDS transforms into acute myeloid leukemia, which has a very bad prognosis.
Allografts, where the stem cells come from a sibling or volunteer donor, have previously been too toxic for most patients with CLL (40% dying from the transplant), but recently mini-transplants, with less intensive conditioning chemotherapy, have proved safer. This is still an experimental treatment and we must wait to see how successful it will be, and how toxic.
This sounds like a silly question. After all, chlorambucil has been around for more than 50 years and is as cheap as chips.
But the truth is that no treatment has been shown to prolong life for longer than chlorambucil.
The only way to satisfactorily compare two treatments is by a head-to-head, randomized, controlled clinical trial of sufficient power to distinguish between the treatments. Many trials that are conducted are just too small to detect differences. The smaller the difference between two treatments, the larger the number of patients needed. When there are a number of small trials, statisticians can sometimes add their results together to obtain a valid answer. This is called a meta-analysis.
There have been randomized controlled trials (RCTs) comparing chlorambucil with CHOP. CHOP is no better.
There have been RCTs comparing chlorambucil with fludarabine. Fludarabine produced more responses, both partial responses and complete responses, than chlorambucil. In addition the lengths of the fludarabine remissions or the progression-free survivals were greater for fludarabine than chlorambucil. Unfortunately, overall survival is no better for those having fludarabine than those having chlorambucil. Supporters of fludarabine say that this is because those treated with chlorambucil were allowed to have fludarabine second time around when the chlorambucil failed, and certainly more patients switched this way than the opposite way. However, supporters of chlorambucil say that if fludarabine works just as well the second time around, why not give chlorambucil first and spare both the expense and the side effects. Furthermore, they say that the dose of chlorambucil was too small, and there is certainly evidence that chlorambucil works better at higher doses.
Three meta-analyses are being conducted, but these have not so far given a consistent answer. Still, there are still similar RCTs that have not yet reported, so we may yet have an answer.
RCTs like this are called phase III trials. When there is no comparison, except with what went on in the past (so called historical controls) the trials are called phase II trials. The MD Anderson Cancer Center is a great champion of phase II trials. For CLL they have done large phase II trials of fludarabine combinations. Two in particular are important: fludarabine and cyclophosphamide (FC) and FC plus rituximab (FCR).
The phase II trial shows a higher response rate (both partial and complete responses) for FC than in a previous phase II trial for fludarabine alone. Similarly, a phase II trial for FCR showed a higher partial and complete response rate than in the previous FC trials. These trials do have overall survival data. And this gives us more things to chew on.
We have to remember that we are dealing with patients in these three trials who had relapsed or refractory CLL. This means that they had been treated previously and either failed to respond or had relapsed after treatment. The median survivals for the three treatments were compared. “Median” is a particular form of “average”. It means in this case the time at which 50% were dead and 50% still alive. On the results reported at the American Society for Hematology (ASH) meeting in 2003, the median survival was 19 months for fludarabine, 29 months for FC and more than 42 months for FCR. The difference between these values was statistically significant.
However, the trials are not comparable. Although they all took place in the same hospital, they did not take place at the same time, so other things might have changed over time – better antibiotics, better anti-fungal drugs, better anti-virals. Moreover, because the patients weren’t randomized between the various types of treatment we do not know whether they were similar groups. All these studies do is provide justification for an RCT.
There are several ongoing RCTs comparing fludarabine with FC. In general preliminary reports have suggested more complete remissions with FC. These trials will probably report in 2005. There are also trials comparing FC with FCR in both untreated and relapsed CLL. When these trials report (probably in 2007) we will have a better idea of any potential benefits.
In the absence of an RCT comparing fludarabine with the combination of fludarabine and rituximab (FR) the CALGB has made a comparison between two of its RCTs which were organized sequentially. From a trial comparing fludarabine with chlorambucil with a combination of both, they took the fludarabine arm and compared it with the combined arms of a trial comparing rituximab given at the same time as fludarabine to rituximab given afterwards. The second group had a significantly longer progression-free survival and overall survival. A multivariate analysis suggested that the only significant difference between the two groups was whether or not they had received rituximab. This is not a randomized controlled study, and it is entirely possible that the two groups differed in some way that was not examined in the multivariate analysis. For example, CD38, ZAP-70, VH gene mutations and chromosomal aberrations were not tested for.
Effective drugs have side effects.
Fludarabine not only kills CLL cells, it kills T-cells. Patients who have been treated with fludarabine have prolonged suppression of their CD4+ T-cells. The levels may go as low as they do in AIDS and remain there for at least 2 years. Patients are therefore susceptible to the same sort of infections as are seen in AIDS. Pneumocystis carrinii pneumonia must be protected against by taking Bactrim (or Septrin) prophylaxis. Attacks of Herpes Zoster (shingles) occur and some doctors recommend valcyclovir prophylaxis. Oral or esophageal candida may demand fluconazole. I have seen reactivation of cytomegalovirus necessitating gancyclovir, and infection with aspergillus which required ambisome.
Fludarabine triggers hemolytic anemia. Chlorambucil or even radiotherapy can do the same, but after fludarabine the hemolysis is very severe and often fatal.
I mentioned earlier that Richter’s syndrome seems to be getting commoner. It now appears that as many as a half of cases of Richter’s syndrome are not caused by a transformation of the CLL, but the development of a separate aggressive lymphoma. Many of these seem to be in response to a reactivation of Epstein Barr (EB) virus, which we all have but which we keep quiescent under the control of our T-cells. When this uncontrolled outbreak of EB occurs the infected B-cells become an aggressive lymphoma.
The worst and least treatable form of CLL occurs when the cells develop a loss of the p53 protein. This protein protects the integrity of the DNA. When it ceases to function, a cell starts to accumulate more and more genetic errors. Cells deficient in p53 are resistant to fludarabine and fludarabine combinations. It seems likely that patients with CLL often have small numbers of cells that are missing p53, but these are kept under control by the presence of the other CLL cells. After effective treatment, even if there is a complete remission and no evidence of minimal residual disease by a sensitive test, there are always some CLL cells left behind. These are the resistant ones; often the ones that lack p53. When they grow back most of the cells will lack p53 and be resistant, and only sensitive to Campath or high dose prednisolone. Campath is even more immunosuppressive than fludarabine and high dose prednisolone not much better.
Disclaimer: The content of this website is intended for information only and is NOT meant to be medical advice. Please be sure to consult and follow the advice of your doctors on all medical matters.
Copyright © 2002-2007 CLL Topics, Inc. All Rights Reserved.
All materials contained on this site are protected by United States copyright law and may not be reproduced, distributed, transmitted, displayed, published or broadcast without the prior written permission of CLL Topics, Inc. You may not alter or remove any trademark, copyright or other notice from copies of the content.
However, you may download and print material from CLLTopics.org exclusively for your personal, noncommercial use.