Chemotherapy combinations

Whenever an oncologist finds two drugs that are both effective in a particular cancer, his natural inclination is to combine them. It makes little sense to combine similar drugs, which work in the same way and have similar side effects. There would be little point in combining chlorambucil with cyclophosphamide or cisplatin with carboplatin or vinblastine with vincristine, but where you can mix drugs that have very different modes of action and different side effects then it often makes sense. One of the most successful combinations has been cyclophosphamide, vincristine, adriamycin and prednisolone (CHOP). Recently it has been shown that adding rituximab to this combination - R-CHOP - is even better in diffuse large cell lymphoma. However, it cannot be assumed that combination chemotherapy will necessarily be better than single agents with every tumor types. A good example of this is the comparison of CHOP and chlorambucil in CLL. Trials have shown that neither is superior, though chlorambucil is less toxic.

The CALGB trial that compared fludarabine with a fairly low dose of chlorambucil also originally had a third arm, fludarabine plus chlorambucil. The dose of fludarabine was reduced in this arm from 125 to 100 mg/sq m/month and the dose of chlorambucil was reduced from 40 to 20 mg/sq m/month. However, recruitment to this third arm was abandoned mid-trial because the toxicity was too great. Infections were more than three times as common as in the chlorambucil arm.

Cyclophosphamide is reputed to be less marrow toxic than chlorambucil, though perhaps more immunosuppressive. It didn't take a rocket scientist to think that the benefits of combination might be allied to less toxicity by combining fludarabine and cyclophosphamide (FC). The MDACC group performed a phase II trial with this combination (published in J Clin Oncol in 2001) with encouraging results. Consequently, both the Germans and the British have performed randomized phase III trials comparing fludarabine with FC. The trials differed slightly. The German trial was restricted to patients under 65, while the British trial also included an arm in which chlorambucil was given at a higher dose. The Germans used a slightly higher dose of fludarabine 150 versus 125 mg/sq m/month in the single agent arm and 90 versus 75 mg/sq m/month in the FC arm. For the latter two thirds of their trial the British used the drugs orally in equivalent doses. The CALGB trial had use 100 mg/sq m/month of fludarabine in the fludarabine plus chlorambucil combination.

Both the British and the German trials showed a higher response rate, higher complete response rate, and longer progression-free survival for the FC arm, but no difference in overall survival. In both trials FC gave significantly more bone marrow toxicity and in the British trial but not the German one, this translated into significantly more hospital admissions. In both trials the incidence of hemolytic anemia was much less in the FC arm.

In the follow up of the CALGB trial a paper published in J Clin Oncol in 2002 reported an excess of patients with myelodysplastic syndrome or acute myeloid leukemia occurring in patients who were entered into the abortive fludarabine plus chlorambucil arm There were 3.5% in this arm compared with 0.5% in the fludarabine arm and 0% in the chlorambucil arm. Indeed Guillaume Dighero, who ran the French CLL trials for many years, told me that he had never seen or heard of a case of MDS or AML occurring after chlorambucil treatment in any of the French trials. The possibility of late side effects after the combination of fludarabine plus an alkylating agent is something that needs to be worried about, and in that respect it is interesting that so far there has been only one patient developing MDS so far in the CLL4 trial. As might be expected this patient received FC, but it is far too early to draw any conclusions.

To sum up: FC might be better than chlorambucil for CLL, but it performs better only in respect of surrogate markers. It does not make you live longer than good old chlorambucil.

Interestingly, NICE looked at the the FC combination and were on the verge of recommending it on health economic grounds - a QALY would cost less than £3000. However, there is no marketing authority for the use of fludarabine in combination in the UK, and because of this NICE declined to make a recommendation. This seems to me to be crazy. There used to be a TV program in Britain called That's Life. Every week they would make a Jobsworth award for the silliest use of petty rules to avoid doing something sensible (Jobsworth comes from the expression "It's more than my jobs worth to allow you to do that"). It was suggested that a Jobsworth award should go to the British Airways official who prevented a check-in worker from wearing a small cross around her neck. With 100 British MPs vowing to boycott Brotish Airways flights, this is a supreme case of shooting yourself in the foot.

I suggest the Jobsworth award to NICE for their refusal to pronounce on the FC combination.