Update on FCR

There has been a great kerfuffle about FCR recently. I have to admit that I was very dubious about this drug combination when it first came out. It was extremely popular in America following its introduction by MD Anderson Cancer Canter (MDACC), and produced higher response rates than any previous regimen. However, MDACC doesn't do phase III randomized controlled trials so it wasn't clear whether FCR was the best thing since color television or whether there was special pleading going on.

Looking at the clinical trials it seemed to be the case that the entry criteria weren't particularly stringent, that patients were younger than normal and often of early stage. Moreover, because the patients were sometimes only seen once at MDACC, then the subsequent courses given by local physicians, the long term consequences of FCR were not particularly well documented.

I have long been wary of fludarabine. I was the one who first recognized the severe autoimmine hemolytic anemia it can trigger and our group was one of the first to recognize that rarely it can cause myelodysplastic syndrome. Indeed I wrote an article in Nature Clinical Oncology, a little while ago, which was entitled FCR: No Country for Old Men. This was in response to a summary paper from MDACC on their experience with FCR. I quote from my paper:

"A large, nonrandomized, phase II trial of FCR in patients with CLL was performed at the MD Anderson Cancer Center.6 In this trial, 300 patients were treated every 4 weeks for a planned total of 6 courses of FCR (rituximab 375–500 mg/m2 on day 1, fludarabine 25–30 mg/m2 daily, cyclophosphamide 250–300 mg/m2 daily on days 1 to 3 of each course and days 2 to 4 for the first course only). The patients were followed up for a median of 6 years, and the complete response rate was 72%—far higher than for previous treatments such as fludarabine plus cyclophosphamide
(35%)7 or fludarabine alone (29%). In CLL, a complete response according to the 1996 National Cancer Institute criteria9 allows for a bonemarrow lymphocytosis of 30%, which might represent considerable residual disease. Consequently, methods for the detection of minimal residual disease have been developed that use either flow
cytometry or polymerase chain reaction (PCR). Tam and colleagues used both techniques and found that 35% of patients were negative for residual disease."

"Although their PCR technique achieved the new international standard of detecting
one tumor cell in 10,000, other researchers have described and used flow cytometry
and PCR techniques that are 5–10 times more sensitive. Among patients with a partial response or better (n = 285) the median time to progression was 80 months, with a projected 60% progression-free survival at 6 years. Again, this result is
impressive but it must be seen in the context of the characteristics of the patients treated. Firstly, the patients in this study were relatively young; the median age was 57 years, and only 14% were aged 70 years or older, whereas in the general CLL population, 70 years is the median age of presentation. As a group, patients over the age of 70 years were significantly less likely to complete the optimum six cycles of therapy (46% versus 79%; P <0.001) and significantly less likely to achieve complete remission (P = 0.02). Secondly, most modern prognostic markers such as ZAP-70 expression, IGHV mutational status and fluorescence in situ hybridization were not studied in this trial. CD38 expression was assessed, but only 21% of patients had more than 30% of cells that expressed this marker. The corresponding
percentage of patients treated in our center is 64%."

"Lacking a contemporary comparator group, the main interest in this trial is the long-term safety profile of the FCR regimen. The actuarial risk of Richter syndrome was 2.5% at 6 years – no more than for historical series. Likewise, the actuarial risk of myelodysplastic syndrome was only 2.8% at 6 years. The well-established effect of fludarabine in depleting circulating T cells is certainly responsible for the 10% risk of serious or opportunistic infections during the first year of remission and 4% risk in the second year. Four of the five late deaths from infection, however, were caused by bacteria rather than fungi or viruses."

"The most unexpected toxicity was persistent cytopenia after completion of therapy, which continued for at least 3 months. This effect occurred in 19% of patients; however, following recovery of blood counts, recurrent late cytopenic episodes occurred in 28% of patients, predominantly in the first year of remission. The use of salvage therapy for patients with persistent cytopenias who relapse is particularly demanding, both for the patient and physician."

Despite my reticence there is now clear evidence that FCR is a superior regimen in CLL. This comes from the German CLL8 trial, which still has not been published in a peer reviewed journal, but was presented at ASH in December 2009. This is how I reported what it found:

The German CLL8 trial was a large one, involving 817 patients with CLL that needed treating. They were randomized to either FC or FCR in standard doses. The group receiving FCR had a higher response rate (95.1 vs 88.4%) and more complete remissions (44.1 vs 21.8%; p<0.001) and a longer progression-free survival 51.8 months v 32.8 months (p<0.001). We've seen this sort of result before, but do the patients live any longer? Note that the follow up is relatively short (median just over 3 years). The Overall Survival rate was 84.1% in the FCR arm versus 79.0 % in the FC arm (p=0.01).

They were able to do a multivariate analysis to look for what factors predict poorer survival. Several factors acted as independent prognostic factors for both progression-free survival and overall survival, including age, sex, FCR-treatment v FC treatment, receiving fewer or more than 3 cycles of treatment, response, 17p-deletion, serum levels of thymidine kinase and ß2-microglobulin, and mutational status of the IGVH genes.

Adding rituximab seems to lead to more neutropenia, but this does not lead to more infections. I presume that the neutrophils are consumed by the CD20/anti-CD20 immune complexes. In fact there were more deaths in the FC arm (86/396, 21.7% versus 65/404, 16.1%). Most commonly death was cause by progressive disease (FC 48/86, FCR 33/65), but there were also more deaths from secondary malignancies in the FC arm (13/86 v 5/65). Of course, in this age group there were also deaths from unrelated causes. Treatment related mortality was just 2.0% in each arm.

One strange finding was that the benefit of FCR did not reach statistical significance for stage C patients (Rai stages III and IV). Why this was so is not clear, but it might be due to the relative immaturity of the study or it may be that these patients, with a higher tumor load, need more intensive treatment. Since platelets falling to below 100 or HB falling below 10 are indications for treatment, perhaps a higher threshold for starting treatment should be adopted.

To summarize: FCR is the only regimen that has been shown to lengthen life over what was previously believed to be the gold standard treatment. However, there are other acceptable treatments that have never been compared head-to-head with FCR. These include FR, FCR-lite, Bendamustine-rituximab and PCR. It is not known whether substituting ofatumumab for rituximab would be an improvement.

It is clear that FCR is too toxic for some patients, and in the UK the combination of chlorambucil and rituximab has recently been shown to be very well tolerated with a high response rate.