The CORRECT Study: Regorafenib vs Best
Supportive Care in Patients With Colorectal Cancer Who Have Progressed on
Standard Therapy
Thomas H. Cartwright, MD:
For oncologists, finding more effective therapies for patients with advanced, metastatic colorectal cancer that has progressed on standard therapy is a high unmet clinical need—the vast majority of patients with metastatic colorectal cancer receives only palliative care. We all see patients who fail first-, second-, and third‑line therapy. These patients have few options for salvage therapy other than a clinical trial.
For oncologists, finding more effective therapies for patients with advanced, metastatic colorectal cancer that has progressed on standard therapy is a high unmet clinical need—the vast majority of patients with metastatic colorectal cancer receives only palliative care. We all see patients who fail first-, second-, and third‑line therapy. These patients have few options for salvage therapy other than a clinical trial.
Regorafenib is an oral agent similar to sorafenib but
targets a wider, slightly different group of tyrosine kinases including
angiogenic kinases (VEGFR1-3, TIE2), stromal kinases (PDGFR-β,
FGFR), and oncogenic kinases (KIT, PDGFR, RET). Grothey and colleagues[1] presented
results from CORRECT, a large, randomized, phase III trial of regorafenib 160
mg/day vs best supportive care in 760 patients with metastatic colorectal
cancer that progressed on standard therapy. Slightly more than 50% had KRASmutations,
and 60% had received 4 or more previous lines of systemic therapies. All
patients had previously received bevacizumab. Treatment cycles were 3 weeks on
and 1 week off. Encouragingly, regorafenib produced statistically significant
benefits in the endpoints of overall survival (OS) and progression-free
survival (PFS).
The improvement in PFS was fairly modest, a median of
two tenths of a month. However, that does not accurately reflect efficacy as
the curves spread out and show a significant difference after the median—the
hazard ratio (HR) for PFS was .049 (P < .000001) (Figure 1). More
importantly, the median OS benefit was approximately a month and a half (5.0
months with placebo vs 6.4 months with regorafenib), which was significant (HR:
0.77; P = .0052). Disease control (a few partial responses
plus stable disease) was achieved in 45% of regorafenib-treated patients vs 15%
of those who received placebo (P < .000001).
In the oral presentation, there appeared to be a
subgroup of patients who experienced longer benefit, but at this point, we do
not have any way of identifying those patients.
Toxicities were relatively tolerable; although many
patients experienced grade 1/2 toxicities, there were few grade 3/4 events. However,
grade 3 hand-foot reaction was seen in 17% of patients receiving regorafenib vs
0.4% of those receiving placebo. Grade 3 fatigue, hypertension, diarrhea, and
rash were also seen with regorafenib but in fewer than 10% of patients.
Alan P. Venook, MD:
One caveat is that this was designed to be a randomized, double‑blind trial, but there was enough toxicity with regorafenib that it could not be effectively blinded. In other words, I think that the patients and doctors knew who was receiving the drug and who was receiving placebo. Due to toxicity, approximately 10% of the patients stopped treatment with regorafenib, even though presumably they were benefiting or, at least, remained alive and without progression. Therefore, although I concur there is an unmet need, my concern is that additional toxicity may be seen, and I am not sure how popular it will be with patients who are receiving supportive care.
One caveat is that this was designed to be a randomized, double‑blind trial, but there was enough toxicity with regorafenib that it could not be effectively blinded. In other words, I think that the patients and doctors knew who was receiving the drug and who was receiving placebo. Due to toxicity, approximately 10% of the patients stopped treatment with regorafenib, even though presumably they were benefiting or, at least, remained alive and without progression. Therefore, although I concur there is an unmet need, my concern is that additional toxicity may be seen, and I am not sure how popular it will be with patients who are receiving supportive care.
In addition, there are issues with the study and its
design. As you point out, the PFS benefit was minimal, yet statistically
significant, at 1.7 months vs 1.9 months (P < .000001), but
there was a large separation between the curves after the median was reached. Because
the study met its endpoint at a prespecified interim analysis (1.5 months), the
study was unblinded and patients receiving placebo were allowed to cross over
to regorafenib. This helped to identify a subset that clearly had benefit with
regorafenib. Biomarker analyses of plasma and tissue samples are ongoing and
should be aided by the study’s clear cutoff point where patients either benefit
or do not benefit.
Thomas H. Cartwright, MD:
Of note, the patients who experienced a benefit had a somewhat better performance score.
Of note, the patients who experienced a benefit had a somewhat better performance score.
Alan P. Venook, MD:
One caveat is that this study employed a 2:1 randomization, and the results did not indicate whether the patients’ rate of progression coming on study was the same. Eligibility included progression within 3 months of or during previous therapy, and those are different criteria. Assuming those are matched, then it is a modest benefit. If they are not matched, then it may not even be that much of a benefit.
One caveat is that this study employed a 2:1 randomization, and the results did not indicate whether the patients’ rate of progression coming on study was the same. Eligibility included progression within 3 months of or during previous therapy, and those are different criteria. Assuming those are matched, then it is a modest benefit. If they are not matched, then it may not even be that much of a benefit.
Again, I think it is an advance, but it also may
impede future research because regorafenib is yet another drug that has to be
accounted for in a population of patients where, in my opinion, we ought to be
putting more emphasis on biomarker‑driven decisions (eg, tumor mutations) and
treating patients accordingly. Therefore, I am not sure how much impact this
modest advance will really have.
Thomas H. Cartwright, MD:
My conclusion is that this agent potentially meets an unmet need, although the benefit is relatively modest. I would not be surprised if regorafenib were approved. In future trials, the investigators may look at ways to select patients and minimize the toxicity, and may evaluate different dosing regimens (eg, continuous vs 3 weeks on and 1 week off).
My conclusion is that this agent potentially meets an unmet need, although the benefit is relatively modest. I would not be surprised if regorafenib were approved. In future trials, the investigators may look at ways to select patients and minimize the toxicity, and may evaluate different dosing regimens (eg, continuous vs 3 weeks on and 1 week off).
No hay comentarios:
Publicar un comentario