[Boris Simkovich]

Interesting stem cell transplantation data

by Boris Simkovich on Fri Jul 11, 2014 11:15 pm

While helping to finalize a news article that we're working on for this weekend, I came across some data that I thought might be of interest to regular forum participants.

One of the questions that often comes up about stem cell transplantation is: How much does it really deepen the response to initial therapy?

In other words, how likely is it that someone who achieved a partial response after their induction therapy will move to a very good partial response, or complete response, as a result of the (autologous) stem cell transplant process?

People have cited data on this point before in article comments and forum postings here at The Beacon. However, it's not easy to find such information, and I happened to come across a study this evening that has a very nice table with results that speak to the issue.

The table summarizes results of a randomized Phase 3 study conducted in France that compared two induction regimens:

1. Regular-dose Velcade and dexamethasone (VD)
2. Lower-dose Velcade combined with lower-dose thalidomide and dexamethasone (vtD).

After these induction therapies, all patients went on to receive an autologous (own) stem cell transplant involving, as usual, high-dose melphalan (link to article [PDF]).

Here are the data on the response rates seen during the study at different points in time:

RRs_VDvsvTD.png (28.49 KiB) Viewed 577 times

These data show, as many people would expect, that the stem cell process does deepen responses.

Does that mean everybody should have a stem cell transplant after induction therapy? Absolutely not -- that's not my point in sharing the data. They are not an answer, by any means, to the "stem cell transplant vs. no stem cell transplant" debate.

I'm sharing simply because the trial involved two treatment regimens that are either common (VD) or similar to a commonly used induction regimen (vtD is analogous to RVD / VRD, in that it combines Velcade with an immunomodulatory agent and dex).

And, as I mentioned, the question of whether transplantation really deepens responses has come up regularly.

Here, however, is the interesting twist to these data. Based on the response data, you might expect the vtD regimen to have the best survival results. The vtD regimen has the highest overall response rate and -- arguably -- deeper responses as well.

So if you believe the argument put forward by many who argue that transplantation is good because it deepens responses, then you would expect the vtD regimen to have the better survival rates.

Well, the paper doesn't provide overall survival results. But the progression-free survival results are not what you'd expect. Here they are:

PFS_VDvsvTD.png (12.79 KiB) Viewed 579 times

The difference in the two survival curves is not statistically significant. But the vtD survival curve actually is below the VD curve for much of the graph.

As I said ... interesting.

[Rneb]

Re: Interesting stem cell transplantation data

by Rneb on Sat Jul 12, 2014 6:54 am

First of all, thank you for posting this info, Boris.

Second, are we to analyze the data utilizing the mindset of Mark Twain, re: Statistics?

Or is it (Survival) possibly referable to differences in functional reserves of the available Immune systems, post transplant v No transplant ?

Thanks.

[Boris Simkovich]

Re: Interesting stem cell transplantation data

by Boris Simkovich on Sat Jul 12, 2014 8:05 am

Hi R,

Yes, always analyze research results keeping Mark Twain's perspective in mind ('There are lies, damn lies, and statistics'). Keep that view in mind, particularly, when it comes to studies that are done retrospectively and/or combine data from different studies, where the choice of the studies included in the analysis could have an important impact on the "results" of the analysis.

We thought these results would be interesting to people, however, because they come from a randomized Phase 3 trial, and such trials are generally particularly valuable sources of information.

Also, the results I've shared above are not investigators' interpretations and explanations of the results, but the basic results themselves -- response rates, and progression-free survival rates. Often, it's when you get to the interpretation of such basic results that you really have to worry about Twain's warning.

I tried to indicate this in my original posting when I wrote that the results shouldn't be interpreted as an answer to the early vs. delayed transplant question. They're just basic data from a trial where all the patients did transplants. You have to make some important assumptions if you want to say that these results favor one or the other view when it comes to early transplantation.

[Multibilly]

Re: Interesting stem cell transplantation data

by Multibilly on Sat Jul 12, 2014 9:09 am

Boris,

Great stuff, as always.

My take on the data is slightly different. What it suggests to me is that if you are going to go down the transplant route, that you absolutely want to try to get to select the path that will get you to CR. If you look at the response rates of the two regiments going into the transplant, you will note that VD had a slight edge on VtD in the CR category in 2/3 of the milestones. I then see this same overall edge carrying through in the PFS data. So, I think the key metric to look at in the response rates is not the ORR (which = CR + nCR + VGPR + VR), but rather just the CR number itself. Of, course, the real numbers I would like to see would be the OS numbers, as that is the one that interests me the most.

So, while I'm not a great transplant fan, I would encourage those considering it to try to everything possible to select the path that will get them to CR/MRD-negative status, whether it be through additional cycles or tweaking the drug mix.

The following link is a long transcript of a recent FDA/NCI roundtable that bears this out (see p 37 since it is a very long read).

http://www.fda.gov/downloads/MedicalDevices/NewsEvents/WorkshopsConferences/UCM401699.pdf

".... So this is the effect of MRD of a Day 100 assessment following a stem cell transplant. And there's a clear benefit for being MRD negative.There is an approximately 14 month PFS advantage. I think that effect is to an extent partially abrogated by the fact that patients were half of the patients were receiving further therapy in the form of thalidomide maintenance. So when we look at the overall survival benefit, there's quite striking survival benefit for MRD negativity, and there's an almost two year overall survival benefit for being MRD negative in this context"

The Beacon article below bears out the importance of getting to CR in a transplant setting:

https://myelomabeacon.org/news/2014/02/26/complete-response-transplant-survival-multiple-myeloma/

Getting close (NCR, VGPR and PR) just doesn't cut it in a transplant setting from a PFS and OS standpoint. You want to achieve CR and MRD-negative status.

Of course, I'll reinforce all the caveats that you mentioned in your original post. Thanks again!

[Multibilly]

Re: Interesting stem cell transplantation data

by Multibilly on Sat Jul 12, 2014 11:34 am

I might also add that the VtD arm had a higher proportion of patients (26% versus 15%) with higher risk cytogenetics (del (17p) and t(4;14)). So, this could also come into play with the PFS data?

It's also interesting that the author finishes up the article by making a case for for VtD in light of the superior PFS with VD. He clearly values ORR over PFS or CR.

[Boris Simkovich]

Re: Interesting stem cell transplantation data

by Boris Simkovich on Sat Jul 12, 2014 7:44 pm

Thanks for the feedback, Multibilly. Always appreciate your perspective.

I had not noticed the difference in the shares of patients with high-risk cytogenetics (chromosomal abnormalities) between the two arms of the trial. That's a good point. That certainly complicates drawing any conclusions about depth of response and its impact on survival.

You also raise a good point about the impact achieving CR, sCR, or MRD- status has on prognosis. However, do you really think the difference in the CR rates between the two regimens -- which is never more than two percentage points -- could be expected to play a role here?

More importantly, the vtD regimen has a higher CR rate right before transplant – a key milestone – yet, with all the caveats we've mentioned, the regimen still has a lower PFS.

In any case, the main thing I wanted people to have access to was the table with the different response rates at different points during treatment. The PFS results are also interesting, but the response rates are something people ask about regularly.

One other interesting aspect of the data ... I've not been able to find updated results with, for example, overall survival graphs. If anyone else can find them, let us know.

The unfortunate thing is that this happens more than you might like. As I've been going back recently and looking at data from several older clinical trials, you often find that there will be an initial report, and then no further follow ups.

[Multibilly]

Re: Interesting stem cell transplantation data

by Multibilly on Sat Jul 12, 2014 8:30 pm

Hi Boris,

All your points are warmly taken.

I admit my conclusions on VTD vs. VD with respect to PFS were a bit of a reach given that the CR response stats are so darn close. I guess the point I want to emphasize is that CR might be a better metric to focus on then the ORR numbers in general, when it comes to transplants.

I used to be a bit more cavalier in my thinking regarding the importance of the level of response going into and coming out of a transplant (i.e. maybe it was OK to have a VGPR going into a transplant). But I've since been convinced that CR matters ... and it matters deeply. If I were personally going into a transplant, I'd be doing everything possible to get to CR going in and not settle for anything less ... assuming my body and situation cooperated

Given my newly found insight on the importance of CR going into a transplant, I'm just trying to get my head around what this might mean for the best non-transplant approach to treatment (which is the route that I will go, should I progress).

I used to think that keeping some of the new, recent drugs in reserve was the way to go for NDMM patients. But I'm starting to question that approach and wonder if it might be better to instead try to hammer the hell out of all the clonal strains up front with some of the newer drug combos, as opposed to thinking about those newer drugs being something I can fall back on down the road? That is, I was thinking I might do something like VRd (Velcade + Revlimid + dexamethasone) or PAd (Velcade + doxorubicin + dex) as an initial therapy, but I'm now wondering if I might instead reach for something like an elotuzumab or carfilzomib [Kyprolis] combo upfront. This topic might actually be better served with a new thread.

As far as the lack of followup data on the original trials, it is indeed a sad situation. But I also really can't blame the researchers for moving on to the next big thing (and funding) after buttoning up a trial. It's just really the reality of research, available time and $$$.