Thursday, June 4, 2015

Frontline Survival Benefit for Ibrutinib


Resonate-2, the randomized phase III study comparing frontline ibrutinib against chlorambucil met its primary endpoint of progression free survival (how long you are both alive AND without progression) and ended early.  More importantly, the secondary endpoint of overall survival was improved as well

Meeting the primary endpoint was a virtual certainty.  Everybody knew this study would be positive - but the study had to be done in order to get FDA approval for frontline ibrutinib.  Single agent chlorambucil is a dead dog.  It has served as the control arm that has been beaten up by just about every new drug in CLL in the last ten years (including campath, bendamustine, fludarabine, obinutuzumab, and now ibrutinib).  

Meeting the secondary endpoint is a little more provocative.  This study was built with a crossover design - meaning if you were randomized to chlorambucil and experienced disease progression, they gave you ibrutinib at that point.  If there is an overall survival benefit, it means that early ibrutinib saves lives compared to ibrutinib after progression.  This was also true in the Resonate study comparing ibrutinib to ofatumumab and the Gilead study that led to approval of idelalisib in combination with rituximab.  That is three studies that show that patients die faster when they get ineffective therapy compared to novel agents.  It remains to be seen if that is true with standard  regimens that actually work like bendamustine-rituximab or FCR where the bar will be much higher than chlorambucil.

Presumably this will be presented at ASH if it isn't already published by then.  It will also lead to FDA approval of frontline ibrutinib but that is probably still a ways off and the exact prescribing "label" will determine who can get it in the frontline.  I have no particular insight, but suspect that is still 4-8 months away.

There will be several key details I want to see in the presentation:

1) In this "elderly" subset of patients - how well was ibrutinib tolerated?  In most clinical trials it appears that about 5-10% of patients discontinue drug for poor tolerance.  Age has also been shown to be biggest variable leading to drug discontinuation - and the median age in this study is likely to be a fair bit higher than other studies due to the nature of the control arm

2) How good is good?  Our group helped contribute to the "frontline ibrutinib" data published in the Lancet.  In that cohort of about 40 patients, only one patient experienced disease progression and several others discontinued for intolerance.  In this larger - multicenter study, it will be the largest to date of previously untreated patients to go on ibrutinib.  Is the progression rate still that good?  Are there more progressions or is it still in the low single digits?

3) What happens to patients who discontinue ibrutinib?  In the relapsed / refractory setting, patients who discontinue ibrutinib have been extremely sick and many have died very quickly.  There has been a fair bit of discussion about "ibrutinib discontinuation syndrome."  Some have argued that these patients had already failed every other therapy which is why they died so quickly.  In this treatment na├»ve cohort - that excuse cannot hold.  If the overall survival curve looks like the progression free survival curve - I will be worried.

4) Dose intensity.  Do patients start 3 pills per day and stay on three pills per day without dose reduction or interruption?  The ASCO data just presented really gives me considerable apprehension about big dose interruptions.  As a patient, I would be trying to keep skipped doses to a minimum.  Obviously you may need to do that for surgery etc. but if I had the choice, I would not want to do that early in a treatment course when there are a lot more CLL cells running around.

5) Cytogenetics and high risk markers:  While chlorambucil is such a dog it really isn't worthy of a comparison arm, it will be interesting to look at the subgroups treated on ibrutinib.  While comparing across trials is always dangerous, we have expectations for the outcomes of patients with markers like TP53, 11Q, and IgHV mutated/unmutated.  This will serve as some interesting food for thought with those higher risk populations.

This news release follows two positive randomized phase III trials presented at ASCO for ibrutinib in combination with bendamustine/rituximab and idelalisib in combination with ofatumumab.  These latter two studies are important but may not have as much practice impact as the study discussed above.

With a frontline indication for ibrutinib likely available within a year, it begs the question who to treat with chemotherapy and who to treat with targeted agents.  I've attached some slides from a talk I recently gave on this topic but please note that these reflect my approach and guidelines are rapidly changing in this disease.

Thanks for reading.  To leave a comment, click on the title of this post and it will open up the post in a new window with a comment section at the bottom.

Sunday, May 17, 2015

Take your pills - ibrutinib dosing matters!

When we are investigating new drugs that have never been tested before, we start with what's called a phase 1 study.  Historically, the goal of a phase 1 study was to define the "maximum tolerated dose." In the era of traditional cytotoxic chemotherapy, you knew you had arrived at that dose when patients simply couldn't handle any more - it was just to much. Perhaps they had too much nausea or vomiting, or the liver couldn't handle it anymore, kidneys failed, or some other toxicity made it clear that you had reached the limits of human tolerance.  As researchers, we just hoped we could get the drug levels high enough without causing too much damage.  If we achieved blood levels that we expected to kill the cancer cells without irreparably harming the patient in the process - that was victory.

Many of the new drugs challenge that paradigm.  When treatments effectively target the specific molecular abnormality with a cancer cell we can see considerably more efficacy while at the same time reducing toxicity.  This has led to the concept of "optimal biologic dose."  Instead of pushing the dose to the max, you only increase the dose as far as you need to - often with substantially less side effects than the traditional therapies.

Unfortunately, "optimal biologic dose" is much harder to define than "maximum tolerated dose."  It presumes that we have effective and accurate means of actually measuring what were trying to do.  While it may come as a surprise to many patients, the unfortunate reality is that there is an enormous amount of human judgment as well as a paucity of clear data involved in early clinical trials. Things are not as scientifically certain in early trials compared to the level of data we have later in a drugs like cycle.  Furthermore, we often need to generalize to the larger population from a very small subset of patients that are appropriate for a phase I study.

There is new data regarding the dosing of Ibrutinib that I think is really important to consider.

Most drugs inhibit enzymes in what we call a "reversible" fashion.  This means that the particular target is turned off only when you have high enough levels of the drug in the blood.  Ibrutinib is a little bit different, it is what we call a "covalent" inhibitor.  When you swallow a pill of Ibrutinib it gets into the bloodstream and either quickly binds to the BTK protein or it gets eliminated from the body.  Within just a few hours of taking a pill there is virtually no free drug in the blood.  Instead, it is all bound to the BTK protein - completely shutting down that signaling pathway until the cell makes more BTK.  This is very different than most oral drugs where we are trying to make sure the levels are still high enough right before you take your next dose.

When you had a sore throat as a kid, the doctor always said to be sure to take all your pills so that the bacteria didn't become resistant.  It is virtually a scientific paradigm that exposing bacteria to inadequate antibiotic dosing creates resistance.  The same is probably true in leukemia and lymphoma with some of the new drugs.  A CLL cell that has its B-cell receptor signaling pathway completely inhibited has a hard time escaping that inhibition.  If that same pathway is only partially inhibited however, it will try to find ways to escape.   This is why the discovery of mutations in the BTK protein that confer resistance to Ibrutinib or so important. These can only arise in cells that have survived the BTK inhibitor long enough to figure out how to thrive under the suppressive influences of Ibrutinib.

Dose intensity can be measured in two key ways.  The first is how many days you take it out of how many days you are supposed to take it.  We already know in other chronic leukemias like CML, adherence to Gleevec (imatinib) is the biggest predictor of treatment success.  Heck, it is even true in breast cancer with hormonal agents.  Now it looks like the same is true in CLL.  Sometimes side effects force you to hold therapy - but prolonged drug holds are undesirable.  Patients who had drug holds in excess of 8 days were almost three times more likely to experience a disease progression (link to ASCO 2015 abstract here).  

The second way dose intensity is measured reflects what dose you take daily compared to the "optimal biologic dose."  There was a very compelling presentation at AACR a few months ago  (I am still trying to figure out how to link to the actual poster, but here is link to the session).  The study title was "Population Pharmacokinetic-Pharmacodynamic (PKPD) Modeling of Ibrutinib in Subjects With B-Cell Malignancies" by Poggesi et al. (prize to the first person who figures out how to find the actual poster).  I need to get a little technical for a minute - stick with me.

As I wrote above, ibrutinib has a cool property that is unique compared to most drugs.  Since it covalently (or irreversibly) binds to BTK, we can do a blood draw, isolate the CLL cells, purify the BTK protein, and look to see how many molecules of the BTK protein are "bound" to a molecule of ibrutinib.  This is what we call, "receptor occupancy."  The higher the occupancy, the more drug is bound to the protein, and the more the pathway is shut off.

You can then ask how many people have how much of their BTK protein "occupied" or inhibited by ibrutinib at different doses.  If we set the bar pretty low at 75% occupancy, any dose above 280mg (two pills) is pretty effective.  96% of patients achieve that level of occupancy at any of those doses.  that may seem good but unfortunately, that low bar means that the pathway is only 75% "turned off."  It is more like a dimmer switch on the lights instead of an on/off.  75% occupied means that there is a lot of room for cells to try to discover ways to become resistant.  If you set the bar much higher at 90% occupancy, the standard CLL dose of 420mg (three pills) can accomplish that in 86% of patients but only 75% of patients who take two pills and 53% of patients taking on pill.  In short - dose matters.  You get more complete pathway inhibition with higher doses of ibrutinib.

What we don't really know what level of pathway inhibition is optimal for CLL treatment.  It is tempting to think that 100% occupancy in 100% of subjects might be much better at preventing eventual resistance - but honestly, we do not know if that might mean higher levels of side effects. How could we get there?  Well, perhaps we didn't actually define the "optimal biologic dose" correctly in the original phase I study.  We pretty much stopped escalating dose because it seemed to be doing what we wanted it to do in the relatively small population of patients we were studying.

I would be curious to go back and do a "dose optimization" study to see if we could modify either the dosing schedule or the actual dose taken to see if we could make ibrutinib work better than it already does.  I also worked on another BTK inhibitor that is no longer in development called CC-292.  We never really got it to behave as well as ibrutinib until we started giving it to patients twice daily. Other BTK programs are looking at this as well.  Another way of potentially addressing such drug limitations is by adding a second drug that acts through different mechanisms.  This should ONLY be done in the context of a research study.  We have three such studies within the US Oncology network of sites (and here as well) including the addition of ublituximab (an updated version of rituximab), the combination of a BTK and PI3K inhibitor, or even BTK in combination with one of the new immune checkpoint inhibitors - pembrolizumab. 

Doctors tend to reduce doses when there are problems.  It is how we think about so many different clinical problems, that we tend to assume it is smart decision making.  In my own patients, I am concerned that dose reductions or prolonged dose interruptions may be causing ibrutinib to be less effective than if we can maintain the dose intensity.

To leave a comment, click on the title of this blog post.  It will open up the blog post in a separate window with a comment field at the bottom.  All comments are reviewed by myself before being allowed to be posted - be patient, I'm a busy guy.

Thanks for reading.

Thursday, May 14, 2015

ASCO 2015 - NHL

OK, so yesterday I put up my choices for the best of CLL abstracts at ASCO.  Here is my list for NHL.  I have to say, it seems like a pretty thin selection.  I looked at all 320 abstracts with lymphoma in the title and this is the best I could do.  Only one big NHL phase three study that is really about getting Gazyva approved in indolent NHL - and we can't even see the data just yet.

I am deeply passionate about management of DLBCL in patients above age 70 and hope to see more pharma moving in this direction.  Seems stunningly foolish to me that fully HALF of all DLBCL cases are virtually ignored in terms of age appropriate therapy trials.

Anyhow, here is my list.  Feel free to use the comment section if you think I am missing something really important.

CAR-T therapy in NHL - yet another entity out there that could be called a "game changer" Durability of responses in DLBCL to be key determinant of value

Novel endpoints in FL studies - technical paper about defining a variable you can actually measure in real time to figure out if you are on right track with new approach

R2-CHOP in DLBCL - trial in progress, but I have hopes this will be legitimate change to standard of care.  No data yet.  Fingers crossed.  Going to be a while before this reads out…. like several years.  Watch for R-CHOP vs G-CHOP to be first phase III DLBCL study to challenge status quo.

Impact of weight an obesity on rituximab dosing - American love for cheeseburgers killing us with lymphoma

Molecular changes in BCR correlate with ABC vs GCB in DLBCL - nerdy article, but I like it

DLBCL in the elderly - the greatest unmet medical need in lymphoid cancers

More on elderly DLBCL - median age 70, yet such patients rarely on prospective trials

Yet more on elderly DLBCL - any drug company with NHL drug is foolish to ignore the opportunity here

Bendamustine +/- Gazyva in rituximab refractory - big phase III study.  May prompt FDA.  Curious to see the data.  We have a press release linked here.  Keep in mind, trials to get approval can be very different than trials to change the way we practice.

Idelalisib in Waldenstroms - didn't make the FDA label, but still active.

Clearing the junk in amyloid - ok not normally a NHL disease symptom, but this is just too cool