Those of you who waded through my post yesterday on clonal evolution will be glad to know this is a much more simple one! I am going to break a statistical rule today but am in good company as the information below is based in part on a table found in the UK CLL 2018 guidelines: onlinelibrary.wiley.com/doi...
The rule I am going to break is that you are not really meant to make simple comparisons between studies. A study stands alone has its own inclusion and exclusion criteria and defined population. It is not therefore directly comparable with another study done at a different time. But when we don't have studies directly comparing two treatment strategies we have little choice but to look at the data together from different studies but carefully recognising the limitations of this. Essentially its all we have.
I'm asking the question what response and survival rates do we see with FCR and Ibrutnib given first line. I will ignore all other treatment options as it seems clear that unless a patient is unfit for FCR these are really the only options that are licensed and evidence based.
Let me define some terms and then present the data.
1. RESPONSE MEASURES
ORR (overall response rate) reflects the proportion of patients who experienced some level or response to the medicine.
CR (Complete Remission) The absence of signs and symptoms of the disease. This will include blood tests and nowadays will typically include assessment of MRD. (see the iwCLL guidelines for more details bloodjournal.org/content/13... )
MRD-ve (minimal residual disease negative) This is the absence of detectable abnormal lymphocytes from the blood or bone marrow. Historical studies have shown that a strong response as indicated by MRD-ve was associated with longer survival (see bloodjournal.org/content/bl... )
3. FOLLOW UP LENGTH
Median Length of follow up is the time half of patients were followed up for. In a way this is more important than the planned length of the study as it recognises drop outs.
3. SURVAL MEASURES
OS - (overall survival) The proportion of patients who are still alive at the end of the study
PFS (progression free survival) The proportion of patients who's disease has not relapsed at the end of the study. This is strongly related to another outcome measure you will sometimes see reported: time‐to‐next treatment (TTNT)
PFS and OS are also often reported in terms of 'median' survival in months rather than as a percentage. This is how long it takes before half of the patients have either died (OS) or relapsed (PFS). This can look alarming to patients at first (and as a patient myself it can look alarming to me too!).
Don't read these numbers as this is the longest you can expect to live without a return of your illness or indeed the longest you can expect to live. Understand that for both progression and death there will always be a group of patients who run into problems early on, and there will always be a group of patients who will do way better than the average.
Also in the study results we will look at in most cases people would not have had the option to switch to the newest drugs (e.g. like venetoclax) when their original treatment failed (because the newer drugs were not yet available). Thus, especially for patients treated with FCR we can now expect that if they relapsed they could be given first ibrutinib and then if that failed venetoclax, and so we would expect the survival statistics will be much better than shown below. What is sorely lacking in my view at the moment is survival data for patients given a series of treatments. (If anyone is aware of such data let me know).
The following data is taken from the paper themselves, from the CLL UK guidelines or from the meta-analysis I refer to at the end of this article. I am ignoring comparators for the purposes of simplicity and because the comparators are not considered mainstream treatment options currently first-line.
--------
Ibrutinib
----------
RESONATE‐2 (Burger et al, 2015 and Obrien, 2016)
nejm.org/doi/full/10.1056/N...
bloodjournal.org/content/12...
Median Length of follow up :18.4 months (in initial study) 60 months when open label extension was included.
Older patients without 17p/ TP53 deletion, NB: patients were not well enough for FCR
Overall Response Rate (ORR) 86%
Complete Response Rate (CR) 4% (rises to 18% with a median follow up of 29 months)
Median Progression Free Survival (PFS) Not Reached (i.e. by the end of the study over 50% of patents had not experienced a relapse of their CLL. This was still the case after the open label follow up period also. In fact 92% of first-line patients had not progressed by the five year open label reporting period.
Overall Survival (OS) at two years 98% and at five years 92%
There has also been a real world data collection of first line patients see bloodjournal.org/content/13...
A total of 391 CLL patients were included with a median follow up of 14 months
Overall Response Rate (ORR) was 81.7%
Complete Response: 17.4%
Median Progression Free Survival (PFS) and Overall Survival (OS) have not been reached (PFS and OS at 12 months were 92% and 95% respectively)
In del17p+ patients, PFS and OS at 12 months were 87% and 89% respectively.
Twenty-four percent (n=94) of patients have discontinued ibrutinib after a median time of 6.5 months
14% (n=55) have required subsequent therapy.
-------
FCR
-----------
CLL 8 (Cramer, 2013)
tandfonline.com/doi/abs/10....
Median Follow up: 47 months
Included some 17p / TP53 patients.
Overall Response Rate 90%
Complete Response Rate 44%
Median Progression Free Survival 51·8 months
Overall Survival 87% alive after three years.
CLL10 ( Hallek, 2010 and Eichhorst,,2016)
thelancet.com/journals/lanc...
thelancet.com/journals/lano...
Median length of follow up 27.9 months
Physically fit patients without 17p/TP53
Overall Response Rate (ORR) 95%
Complete Response Rate (CR) 40%
Median Progression Free Survival 55·2 months
The median overall survivals have been looked at in the combined data of both these studies. See ascopubs.org/doi/full/10.12...
What is seen is that for patients who got only a partial response without MRD-ve status the median overall survival was 72 months and that for anyone with a better response than that the median overall survival has not yet been met meaning that over 50% of patients are still alive.
In an age of Ibrutinib and other new treatments it is easy to forget what a good treatment FCR is. Bear in mind that it was used in CLL8 in patients we now know are not suitable for it, and that due to the timings of these studies most of these patients would not have been able to get newer drugs. Despite this, the overall survival figures patients on FCR in these two studies are really rather good. This is good news for all of us as it suggests if you can get a good response to treatment then you will live a long time. And of course if you fail FCR other treatments are available in the modern era.
MORE RECENT FCR studies
Two newer FCR studies have looked at oral treatment with FCR and shown even better results
Howard, 2017
onlinelibrary.wiley.com/ser...
Overall Response Rate: 93%
Complete Response Rate 76%
Median Progression Free Survival 52·7 months
Munir, 2017
Overall Response Rate 97%
Complete Response Rate 69%
Median Progression Free Survival 51·7 months
DISCUSSION and COMPARISON
==============
Notice that all these studies wildly differ in length of follow up and patient type. So for example the ibrutnib study and one of the two FCR studies both exclude 17p deletions / TP53 issues.
Experience from other Ibrutnib studies would suggest that with longer treatment you would get more patients responding and showing deeper responses. Equally you would presumably get more patients seeing disease progression.
--------
Systematic Review and Network Meta-analysis
---------
In an attempt to use clever maths and other studies to compare these data a systematic review was conducted by the makers of ibrutinib, Jansen.
futuremedicine.com/doi/pdfp...
It is important to not the limitations of this work due to how different the studies are. But by using a network approach. This essentially involves clever maths looking at how drug A compares to drug C and how drug B also compares to drug C, and then making an inference about how drug A compares to drug B. The conclusion was that Ibrutinib is probably the best first line drug in terms of response and survival measures. This conclusion should be treated with some caution, however, not least because the newer FCR studies couldn't be included and FCR was used in some patients we now acknowledge may not be best for it.
CONCLUSON
In terms of deciding between FCR and Ibrutinib first-line in patients who are suitable for FCR this data is perhaps not conclusive. In terms of depth and speed of response this would favour FCR, but Ibrutnib fared very well in the network meta-analysis and it looks likely it will see deeper responses if given for longer and that it may well be better than FCR in terms of progression free survival and overall survival. However, whether any benefit in overall survival for using ibrutinib first would still be present if patients who relapse after FCR are later given ibrunib is not clear at all. There is still uncertainty about this question and so the FLAIR trial remains an important study in my view after reading this.
See this post for more on the UK FLAIR study and the site list:
healthunlocked.com/cllsuppo...
What is your take on this data?
This is an unlocked post so please do not share personal medical data you do not want to appear in google.
Ibrutinib Partial Response
Hospital response to lack of monitoring after first FCR
Ibrutinib denied to new patients
