Background: The results of 144 pts with metastatic CRPC treated in a phase II NRE cohort with daily cabo
100 mg and 40 mg starting doses have been previously reported. Substantial rates of bone scan
improvement, reductions in CTC counts and pain relief were observed. To better understand the implication
of these effects, the association with OS was explored. Methods: Relevant baseline variables (LDH, BSLA,
visceral disease, pain, hemoglobin, CTCs) and post-treatment changes at week 6: $30% reduction in BSLA
using computer-aided assessment, CTC conversion (.5 vs. 4 or less/7.5 ml of blood), and pain intensity (7
day averaged worst pain score; BPI scale; using an IVR system) were associated with OS in 144 CRPC pts
with bone metastasis who progressed within 6 months of docetaxel (D) treatment ($225 mg/m2
) in either
bone or soft tissue. Median OS was compared between responders and non-responders for each of the above
outcomes categories using a Cox proportional hazard model. The findings were examined further after
adjusting for significant baseline covariates selected from a stepwise Cox regression model. Results: See
Table. Conclusions: Recognizing the limitations of associating response with survival, this retrospective
analysis of decreases in BSLA, CTC conversions and reductions in pain intensity support further study in
ongoing phase III trials. Clinical trial information: NCT00940225.
Baseline characteristics, N5144.
Median age 66 Sites of disease, %
Prior therapies, % Bone 100
$2 prior lines therapy including docetaxel 73 Visceral 31
Cabazitaxel 24 Moderate to severe pain (BPI$4), % 47
Abiraterone 43 Median CTC count 37
Trying to compare uncontrolled data from an open label trial like this to a theoretical control arm in a clinical trial is a tough deal. Assuming a 7 month survival for the Comet controls was a bit arbitrary to start with. This was the median survival for patients on the COU 301 trial after progressing on Abi. The typical patient may have a gap between discontinuing one therapy and starting another that is not accounted for in that 7 months. We'll see.
Here's what the HR's (hazard ratios) mean. Over any given time segment, patients who either convert below 5 CTC's, have a bone scan response, or a pain response suffered a mortality risk at about half the rate of their counterparts who did not enjoy any of those benefits. When one considers that historically 2/3 of Cabo treated patients have a bone scan response, 1/2 have a pain response and 40% convert to CTC
I see your point when it comes to reading out the specific results of COMET-1 as it's designed, and in addition to not having necessarily corresponding groups for the calculation there needs to be a proportionality assumption on the Kaplan-Meier curves for the hazard ratio to really mean something. Otherwise, integrating each hazard over a time period wouldn't give you an appropriate interpretation of the accumulation of overall hazard over two groups. I hope Scher has a nice graph to show us.
The result is muddled at best because of dosing and pre-treatment profiles, but it doesn't seem negative. With the tolerability of lower dosages, think it's perfectly reasonable to treat those who respond to Cabo with escalating dosages and take other patients off. As for EXEL, don't forget Cabo's other indications and GDC-0973.
Freakin Yahoo. When one considers that historically 2/3 of Cabo treated patients have a bone scan response, 1/2 have a pain response and 40% convert to CTC less than 5, there is some good news here. None of the Comet controls will enjoy those benefits. Offsetting that, none of the controls will suffer treatment associated toxicity either. If one is trying to handicap the Comets, one has to weigh the potential benefits against the treatment related detrimental factors.
Hazard ratios of responders to non-responders not adjusted for co-variates (left) and adjusted (right) by category of response, 95% interval in hyphens, p-value:
Bone Scan: 0.62, 0.38-1.00, 0.054 0.47, 0.28-0.79, 0.005
CTC Conversion: 0.40, 0.21-0.78, 0.007 0.42, 0.19-0.92, 0.031
Pain: 0.65, 0.34-1.24, 0.186 0.51, 0.24-1.11, 0.090
A possible issue is that if Cabo worsened mortality in non-responders, these figures will be exaggerated; but all in all I'm relieved to see this.
It's not going to help the stock price. 10.8 months compared to the anticipated 7 months for the Comet control groups is not much of a margin especially when it is realized that the Comet control arms will have had more lines of treatment than these patients. Those patients who had CTC conversions, bone scan responses and pain responses did better than their counterparts who did not enjoy those benefits.
"It's not going to help the stock price. 10.8 months compared to the anticipated 7 months for the Comet control groups..." Ernie, I am not quite understand this, please help me, On ALSYMPCA overall survival: radium-233 had medium OS:14.0 months versus placebo OS: 11.2 months. we all know what happen today. Xofigo got approval. why 10.8 months versus 7 months on Cabo is not good?
Sentiment: Strong Buy
Sorry. Considering the refractory patient pops, I like what I see.
All these patients were post-docetaxel...half were post-Abi or post Enzalutamide...
Half were heavily treated w/ existing approved bone targeted therapies, including a single percentile of RA-223. How many halfway-in -the -basket patients do you expect to save? I hate to be this crude, but this is a very sick, heavily pre-treated population we are addressing...and the statistics should bely this fact. For my own part, I'm not taking any action till I hear the presentation. This stock has been overplayed and over-manipulated by the financial community for too long, and it's time to truly pay attention to the science to see where this is going. The fact is that the shorts are departing the trade - slow that it may be - I'd pay attention to those numbers.
I'm still convinced that the shorts want your cheap shares, and will contrive any means to steal them from you...and steal is appropriate word.
From a brief glance, hazard ratios look quite good for responders in each category of Bone scan response, CTC conversion, and Pain compared to non-responders. I'm not certain how "non-responders" will correlate to patients in a placebo group, but my hunch is that HR will be well under 1 with a high level of confidence. Their conclusion is qualified but positive.