Tuesday, November 26, 2013

Off label drug use

How do you get access to a hot new drug if it isn't approved in your condition?  With recent or pending approvals of ibrutinib, idelalisib, lenalidomide, obinutuzumab, etc. lots of people are asking these questions.

In the past, a drug that got approved by the FDA could largely be used by oncologists without much regard to the the specific "labeled indications"  (see ibrutinib / imbruvica here) When a clinical trial is done that leads to the approval of a drug, the FDA evaluates the data and creates a "label."  That label defines what the drug is approved for and that is generally based upon the specifics of the study that led to the approval (ie. the specific population, what other drugs were used, etc).  There is a lot of attention to how broadly or narrowly a label is written because it will have enormous downstream impacts on utilization.

When brentuximab was approved in Hodgkin's lymphoma, it was originally approved for patients with relapse following stem cell transplant, or for patients ineligible for transplant who had received two prior regimens.  Some were hoping the label would be more inclusive (ie to get patients to transplant, or in first relapse etc).  That is why the stock initially went down substantially immediately after approval. Since it was a more narrow label doc's couldn't go prescribe it as frequently as some had hoped.

Many of the same questions surround ibrutinib.  It was given "breakthrough designation" for patients with relapsed CLL that has the 17P deletion.  Recall that 17P is one marker for high risk CLL among several others that determine prognosis.  If I read the press releases correctly though, pharmacyclics submitted for approval in patients with relapsed CLL (irrespective of the 17P deletion).  That sets up a potential scenario where the specifics of the label can make a huge difference in the initial approval in CLL.  In relapsed CLL patients with 17P represent only about 20-30% of all patients.  So if you don't have that marker will you be able to get the drug?  What if you have a P53 mutation without the deletion?  That is essentially the same population?  How will insurance make that decision?  If the drug is approved for relapsed disease only, what would stop a patient from taking a single dose of rituximab and then saying their disease is "previously treated?"

So who governs what can or can't be prescribed?  That is complicated.  For patients with medicare, it is hard to prescribe any of the "new" medications "off label."  This is because medicare won't pay for it.  For some of the older drugs though, they really got established into disease management before they became super expensive.  If a drug gets "NCCN compendia listing" (which is an association of leading academic medical centers who agree on standards of care) it often gets covered by medicare.  Nowadays it is pretty hard to get "compendia" listed unless a specific trial proves that a drug has a role in a very specific situation.

It is possible to prescribe a drug and administer it, but if it goes for ten thousand dollars per dose and medicare won't reimburse you for it, you can put yourself into enormous financial pain very quickly as an oncologist by getting too adventurous.

For patients with private insurance, it can be a lot more difficult to predict.  If ibrutinuib or lenalidomied are currently approved for mantle cell lymphoma, what would stop you from prescribing them in CLL? Since the drugs can cost 10K/month, getting insurance to pay for it is the biggest issue.  These drugs have been clearly well studied in these diseases, it is not like a doc would just be "flying by the seat of their pants" to recommend such a drug.  The reality is that some insurance will cover it and some won't.  In the case of ibrutinib, it will likely be approved in CLL in a few months so it may not be as much of an issue, but for a patient who needs therapy for CLL NOW and wants ibrutinib it all comes down to whether insurance will pay for it or not.

Since the doses are a little different between the two diseases, it shouldn't be a surprise that insurance could figure it is being given "off label."  When you submit a "prior authorization" to insurance for a drug, they also ask for a diagnosis code (ICD-9) that is specific to the disease - i.e. CLL vs MCL vs other.  If the code and the drug don't match, that is another red flag.

Does that mean it can't be done?  I've had success giving patients unusual drugs off label where it was clear to the patient, insurance, and everyone else that we were off label.  In some cases there was something uniquely compelling such as a specific mutation that matched a specific drug in a patient who had received all the available therapies already.  In short - there really are not rules - just financial realities.  In the UK, Europe, and Canada, it is a lot more strict and "off label" use is a lot less frequent. I guess our reputation as cowboys in the US has some truth.

For patients who are determined, a careful, thoughtful letter to the insurance company by the patient or the doctor citing relevant medical literature can sometimes help.  I would suggest strongly that a respectful dialogue goes a lot farther than yelling and screaming in such a situation.

With a bunch of new drugs coming, the landscape of CLL and NHL are going to change dramatically. It will move fits and spurts.  With all the dramatic changes coming in health care (that I have written and tweeted about quite a bit)  it is really hard to predict how it is going to play out.

Thanks for reading