Managing metabolism and drug toxicity in acute lymphoblastic leukaemia

Dr Mary Relling - St Jude's Children's Research Hospital, Memphis, USA

I will be talking about the toxicity of drugs that are used to treat childhood ALL, as well as drugs that are used to treat adult ALL, and I’ll be talking about how, of course, one of the reasons behind toxicity to ALL medications lies in the fact that we have inter-patient variability in response to medications. Some patients respond exactly as we hope with efficacy and no toxicity but there’s always a percentage of patients who have toxicity. So part of what we try to understand is what are the mechanisms behind that inter-patient variability, what are the mechanisms behind why patients have certain toxicities to each specific ALL drug.

There are many, many reasons why there’s inter-patient variability, those include age, diet, liver function, renal function, drug interactions, environmental exposures and, of course, underlying it all is genetic variability. So what we try to do is elucidate those mechanisms for variability and response to drugs, do what we can to minimise that variability and avoid toxicity while not compromising the desired anti-leukemic efficacy for ALL.

So I’ll give a couple of examples. One example for drug interactions is the fact that anticonvulsant medications induce drug metabolising enzymes and therefore they increase the clearance of the anti-leukemic drug. So we and others try to avoid the use of enzyme inducing anticonvulsants concurrently with ALL therapy to decrease the chance for that drug interaction. Also I’ll mention briefly how renal function can affect toxicity. Renal function is really important for the clearance of a drug like methotrexate and at St Jude we use high dose methotrexate to treat ALL but we always target the plasma systemic exposure of methotrexate by estimating each patient’s pharmacokinetics during the drug infusion, changing the dosage during the infusion to achieve an optimal targeted systemic exposure. By doing that we can maintain the drug exposure that we need to treat leukaemia but minimise the chance of toxicity from methotrexate.

We use genetics a lot, germline inherited genetics, to target the way that we prescribe medications for all of our children with leukaemia. There are a couple of genes that are specifically important for treating leukaemia. One of them is thiopurine methyltransferase and my colleague, Jun Yang, showed recently that another really important gene is NUDT15. So those variations that are inherited in those two genes have a huge impact on the toxicity of the thiopurine drugs, 6-mercaptopurine and thioguanine. So what we do is up front we do germline genetic testing for all of our patients to identify what gene variants they have that put them at risk of toxicity and we pre-emptively target the dosages of their medications, especially their thiopurine medications, to avoid that toxicity at all. We do that also for a lot of the supportive care drugs that are used to treat ALL, so rasburicase is used to prevent hyperuricemia and we test for G6PD deficiency. Voriconazole is an antifungal that’s used to prevent fungus infections and we dose that drug based on their CYP2C19 genetic status. So we have an array of genes that we test in every single child and then we use that to adjust how we prescribe therapy.

Then I’m going to end by talking about the drug asparaginase which is used for childhood ALL and for adult ALL. One of the most common side effects or toxicities of that drug is allergy. It’s a non-human protein and so patients develop allergy to asparaginase. We’ve recently determined how to test for antibodies in the serum of patients that are treated with that drug. Patients who have real allergies to asparaginase do have antibodies not only to the protein but also to its conjugate, the polyethylene glycol that it’s conjugated to. So we can figure out which patients are truly having immune reactions and use that information to decide who it’s possibly safe to re-challenge in and who it’s not safe to do that re-challenge and in whom we might have to find an alternative agent. So I’ll briefly hit on all of those topics in my 15 minutes.

Is genetic testing used routinely in clinics at the moment?

Worldwide germline genetic testing is not done routinely. Some of the most recent surveys in the US indicate that fewer than 2% of patients who are prescribed a drug for which the FDA either recommends or requires genetic testing actually have that genetic testing done. That’s actually at a leading academic health centre. So, no, unfortunately even though we know that germline genetics has a big impact on probably about thirty medications that are commonly used, they are very rarely actually tested for.

Why is that? Is it due to cost?

It’s partly due to cost. In many countries the primary driver behind paying for healthcare is treating bad things that happen, there’s much less emphasis on preventing bad things from happening. So many third party payers and insurers are unwilling to do testing that’s preventative testing which is what pharmacogenetic testing can be if it’s done in the most cost effective way possible. At St Jude we have been pre-emptively testing all of our patients since 2011 with a genetic array that tests for hundreds of genes and we place those results in the patient’s medical record as we’re developing the clinical decision support that tells prescribers how to use that information.

Then worldwide I co-lead with Teri Klein at Stanford a group called The Clinical Pharmacogenetics Implementation Consortium. The reason we formed that consortium was that to address one of these barriers that is preventing from genetic testing being used. That barrier is that people don’t know how to use genetic testing to guide prescribing. So we write very specific gene-drug pair clinical guidelines that tell prescribers exactly what they need to do given any patient’s genetic profile for those drugs where it’s important to act. So we hope that that’s going to at least address that barrier behind why some people don’t do genetic testing.

How do you see this developing in the future? What are the next steps?

Unfortunately the next steps are more of the same. They’re what we should have been doing all along which is continue to enrol patients with leukaemia on clinical trials. The purpose of clinical trials is not really just to test out new therapies, it’s to test existing therapies to capture how we actually deliver the therapy that we think we’re delivering and to capture information on what toxicities do occur and to continue to learn what are the best ways to give the medications we have to maintain our desired cure rates and minimise toxicity. And to always couple those clinical trials with things like genetic tests of both the leukaemia and the inherited germline genomic variants so we can continue to learn how to use medications.