Xiaoxiao:

Thanks for your intelligent observations and valuable feedback. In the spirit of advancing the conversation further, I will suggest some qualifications here in support of a more nuanced view.

Correct: it is indisputable that cabozantinib as a multitargeted TKI (in the sense of targeting several points within a single pathway, OR multiple distinct pathways) exerts significant activity against multiple targets including MET and VEGFR2 (as well as the RET, KIT, AXL and FLT3 pathways). But it is equally indisputable - on both preclinical (blocks vascular leakage, effecting vascular normalization, in c-MET positive tumor models), and clinical data (where - by virtue of c-MET inhibition - it has overcome MET-dependent acquired resistance to EGFR inhibitor therapy in a dual c-MET/VEGF inhibition regimen) - that it is nonetheless a c-MET inhibitor (effectively inhibiting c-MET phosphorylation), just not a highly selective one. (And studies have also demonstrated that androgen receptor (AR) signaling represses the expression of MET, which supports future studies of targeting the combination of the AR and MET pathways while in addition, there is a well-established association between VEGF regulation and the activation of c-MET signaling in prostate cancer).

So it is clear that cabozantinib stands among the nonselective c-MET TKIs - cabozantinib, crizotinib (which targets MET as well as ALK, approved for EML4-ALK positive NSCLC), and foretinib (which inhibits c-MET, VEGFR2, PDGFR, and KIT), in contrast to the more target-selective c-MET TKIs such as tivantinib (ARQ197), PF-04217903 (an ATP-competitive highly selective MET inhibitor), EMD 1214063, EMD 1204831, AMG 337, INCB028060 and dozens of others, and we have no systematic robust evidence that selective inhibition provides superior outcome or safety.

As to the relevance of IC50, it's been noted for instance with respect to the c-MET inhibitor tivantinib (ARQ197) that in phase I trials it demonstrated systemic drug levels significantly above its in vitro IC50 values [1], with confirmed c-Met inhibition in pre- and post-treatment tumor biopsies. Until the prediction value of IC50 is robustly validated, it remains at best a crude heuristic, and I have found, one with often rather indifferent translation into the human clinical domain.

But you make an excellent point re the potential value of multitargeting therapeutics, with which I am in wholesale agreement (although on how, there may be some differences). So, the Ann Arbor trial reported at ASCO 2011 demonstrated the cooperative role for c-MET and VEGF signaling in the progression of CRPC, strongly suggesting that dual targeting/inhibition of tumor and microenvironment may favor improved outcome for patients, given that the limited survival impact of VEGF inhibition alone is likely secondary to the development of evasive resistance after inhibition of angiogenesis, possibly through upregulation of MET (hepatocyte growth factor receptor) signalling (given its important roles in angiogenesis, tumor cell invasion and bone metastasis). Furthermore, it's been observed that upon initial diagnosis of NSCLC, MET gene amplification is rare, but rather acquired MET amplification has been noted in up to 20% of EGFR-mutated tumors AFTER pretreatment with an EGFR TKI, and with - in MET gene amplifying tumors - stimulation of the tumor appearing to occur via the co-receptor HER-3 triggering PI3K signaling pathway activation, thereby circumventing the effects of an EGFR TKI. This makes a strong argument for multitargeting therapeutics [2]. It remains to be determined however, I will note, whether the vertical versus horizontal model of blockade will prove superior in outcome (and/or tolerability); namely blockade of serial steps in signaling pathways (vertical blockade), and/or blockade of parallel signaling pathways (horizontal blockade), a continuing controversy.

Finally, I note again the challenging toxicity profiles of c-MET inhibitors, selective or non-selective: thus in the EXAM Trial, most importantly an atypically high 79% of patients had dose reductions, while 16% of cabozantinib-treated patients discontinued treatment as a result of toxicities, and worse, grade 5 lethal toxicity actually occurred in 7% (16 patients) of patients treated with cabozantinib, attributable to the study drug, including from fistula, respiratory failure, sudden death, hemorrhage, sepsis, and cardiopulmonary failure.

THE CHALLENGES IN THE FUTURE

Going forward, it is clear to me that we need to know a lot more about:

(1) the efficacy-optimal and safety-maximal deployment of selective and non-selective c-MET inhibitors, and about

(2) their complex cross-dynamics with both other TKIs and with conventional chemotherapies (as witness the synergy with gemcitabine), and also

(3 the optimal scheduling strategy re vertical versus horizontal blockade (see above), and

(4) and within horizontal blockade, which "array" model to use: either (a) arrays of individual selective TKIs (selective TKI array model), or (b) single, dual or n-tuples of single multitargeted TKIs (parallel TKI array model), and

(5) what the toxicity profiles will be within the deployment of horizontal multitargeted blockade, given the daunting issues of collective toxicities, and

(6) finally, in the wake of the meta-analytic demonstration I discussed above, whether such complex deployments will result in durable and robust significant outcome advantages over conventional chemotherapies, an unexamined dogma for far too long (and one which I predict will increasingly inform FDA, EMA, and other evaluations hereafter).

REFERENCES

1. Gibney GT, Aziz SA, Camp RL, et al. c-Met is a prognostic marker and potential therapeutic target in clear cell renal cell carcinoma. Ann Oncol 2013; 24(2):343-9.

2. Sgambato A, Casaluce F, Maione P, et al. The c-Met inhibitors: a new class of drugs in the battle against advanced nonsmall-cell lung cancer. Curr Pharm Des 2012; 18(37):6155-68.

As you noted, crizotinib (Xalkori) is already FDA approved (2013), and although for ALK-positive locally advanced or metastatic NSCLC, it nonetheless still influences the c-MET pathway, and cabozantinib (Cometriq), a c-MET inhibitor, was approved by the FDA in 2012 for the treatment of metastatic medullary thyroid cancer (MTC), while the experimental c-MET inhibitor tivantinib for NSCLC is now doing well in Phase III trial after modest dose reduction. Over a dozen additional c-MET inhibitors are currently in clinical trials.

However, the toxicity profiles of c-MET inhibitors are challenging: cabozantinib bears a black-box warning re severe and fatal bleeding/perforations and fistula in the colon which occurred in some patients during clinical trials, and this has sobered many oncologists into the appreciation that biological targeted therapies still have significant benefit/harm concerns. Furthermore, there will now be increasing scrutiny of relative therapeutic advantage over traditional chemotherapies, given the dramatic wake up call of just published Korean JAMA meta-analysis showing that median PFS incurred from EGFR inhibitors in advanced NSCLC was inferior to that incurred from conventional chemotherapy, and although this may be EGFR inhibitor-specific, there is no reason to discredit the possibility that other targeted therapies, when critically meta-analyzed as to survival outcomes, may offer only minimal if any advantage over optimal chemotherapeutics, a growing concern now in the oncology community.

@Constantine, thanks for your answer!

As far as I know, cabozantinib is a multi-targeted TKI. Its potency towards VEGFR2 is about two orders of magnitude higher than it towards Met, Ret, Kit and other targets. I might guess the severe side effects of cabozantinib, like bleeding/perforations that you mentioned, probably due to its extreme low IC50 (0.035nM) to blood vessel endothelial VEGFR2. And as for tivantinib, experimental data has indicated that it also targets microtubule, and inhibits cell viability with similar potency in both c-MET-addicted and nonaddicted cells. So in my point of view, both cabozantinib and tivantinib are not specific Met inhibitors.

I am totally agree with your concern about the real advantage of existing targeted therapies over chemotherapies. Since cancer is such a complex multi-factor disease, counting on the inhibition of only one target is hopeless. However, combination of several specific TKI is promising -each inhibitor is highly specific in order to minimize side effects, combinative treatment exaggerates efficiency while lowering the concentration of each single compound to further decrease the toxicity. This is my opinion.

Xiaoxiao:

Thanks for your intelligent observations and valuable feedback. In the spirit of advancing the conversation further, I will suggest some qualifications here in support of a more nuanced view.

Correct: it is indisputable that cabozantinib as a multitargeted TKI (in the sense of targeting several points within a single pathway, OR multiple distinct pathways) exerts significant activity against multiple targets including MET and VEGFR2 (as well as the RET, KIT, AXL and FLT3 pathways). But it is equally indisputable - on both preclinical (blocks vascular leakage, effecting vascular normalization, in c-MET positive tumor models), and clinical data (where - by virtue of c-MET inhibition - it has overcome MET-dependent acquired resistance to EGFR inhibitor therapy in a dual c-MET/VEGF inhibition regimen) - that it is nonetheless a c-MET inhibitor (effectively inhibiting c-MET phosphorylation), just not a highly selective one. (And studies have also demonstrated that androgen receptor (AR) signaling represses the expression of MET, which supports future studies of targeting the combination of the AR and MET pathways while in addition, there is a well-established association between VEGF regulation and the activation of c-MET signaling in prostate cancer).

So it is clear that cabozantinib stands among the nonselective c-MET TKIs - cabozantinib, crizotinib (which targets MET as well as ALK, approved for EML4-ALK positive NSCLC), and foretinib (which inhibits c-MET, VEGFR2, PDGFR, and KIT), in contrast to the more target-selective c-MET TKIs such as tivantinib (ARQ197), PF-04217903 (an ATP-competitive highly selective MET inhibitor), EMD 1214063, EMD 1204831, AMG 337, INCB028060 and dozens of others, and we have no systematic robust evidence that selective inhibition provides superior outcome or safety.

As to the relevance of IC50, it's been noted for instance with respect to the c-MET inhibitor tivantinib (ARQ197) that in phase I trials it demonstrated systemic drug levels significantly above its in vitro IC50 values [1], with confirmed c-Met inhibition in pre- and post-treatment tumor biopsies. Until the prediction value of IC50 is robustly validated, it remains at best a crude heuristic, and I have found, one with often rather indifferent translation into the human clinical domain.

But you make an excellent point re the potential value of multitargeting therapeutics, with which I am in wholesale agreement (although on how, there may be some differences). So, the Ann Arbor trial reported at ASCO 2011 demonstrated the cooperative role for c-MET and VEGF signaling in the progression of CRPC, strongly suggesting that dual targeting/inhibition of tumor and microenvironment may favor improved outcome for patients, given that the limited survival impact of VEGF inhibition alone is likely secondary to the development of evasive resistance after inhibition of angiogenesis, possibly through upregulation of MET (hepatocyte growth factor receptor) signalling (given its important roles in angiogenesis, tumor cell invasion and bone metastasis). Furthermore, it's been observed that upon initial diagnosis of NSCLC, MET gene amplification is rare, but rather acquired MET amplification has been noted in up to 20% of EGFR-mutated tumors AFTER pretreatment with an EGFR TKI, and with - in MET gene amplifying tumors - stimulation of the tumor appearing to occur via the co-receptor HER-3 triggering PI3K signaling pathway activation, thereby circumventing the effects of an EGFR TKI. This makes a strong argument for multitargeting therapeutics [2]. It remains to be determined however, I will note, whether the vertical versus horizontal model of blockade will prove superior in outcome (and/or tolerability); namely blockade of serial steps in signaling pathways (vertical blockade), and/or blockade of parallel signaling pathways (horizontal blockade), a continuing controversy.

Finally, I note again the challenging toxicity profiles of c-MET inhibitors, selective or non-selective: thus in the EXAM Trial, most importantly an atypically high 79% of patients had dose reductions, while 16% of cabozantinib-treated patients discontinued treatment as a result of toxicities, and worse, grade 5 lethal toxicity actually occurred in 7% (16 patients) of patients treated with cabozantinib, attributable to the study drug, including from fistula, respiratory failure, sudden death, hemorrhage, sepsis, and cardiopulmonary failure.

THE CHALLENGES IN THE FUTURE

Going forward, it is clear to me that we need to know a lot more about:

(1) the efficacy-optimal and safety-maximal deployment of selective and non-selective c-MET inhibitors, and about

(2) their complex cross-dynamics with both other TKIs and with conventional chemotherapies (as witness the synergy with gemcitabine), and also

(3 the optimal scheduling strategy re vertical versus horizontal blockade (see above), and

(4) and within horizontal blockade, which "array" model to use: either (a) arrays of individual selective TKIs (selective TKI array model), or (b) single, dual or n-tuples of single multitargeted TKIs (parallel TKI array model), and

(5) what the toxicity profiles will be within the deployment of horizontal multitargeted blockade, given the daunting issues of collective toxicities, and

(6) finally, in the wake of the meta-analytic demonstration I discussed above, whether such complex deployments will result in durable and robust significant outcome advantages over conventional chemotherapies, an unexamined dogma for far too long (and one which I predict will increasingly inform FDA, EMA, and other evaluations hereafter).

REFERENCES

1. Gibney GT, Aziz SA, Camp RL, et al. c-Met is a prognostic marker and potential therapeutic target in clear cell renal cell carcinoma. Ann Oncol 2013; 24(2):343-9.

2. Sgambato A, Casaluce F, Maione P, et al. The c-Met inhibitors: a new class of drugs in the battle against advanced nonsmall-cell lung cancer. Curr Pharm Des 2012; 18(37):6155-68.

Thanks for your professional and informative feedback, Constantine. Have a good day!

Soryy I am new at understanding c-MET inhibition effects. Any comments on its efficacy alone if any  in animal models - particularly the more selective c-MET inhibitors - on vascular leakage; and separately on angiogenesis (without VEGF inhibition).  Thank you.

Does the unfavorable toxicity profile suggest we need to find the right drug combination in order to reduce the dose of a MET inhibitor ? 

From a recent patient forum on inhibitor-resistant lung cancer, the following MET inhibitors were mentioned as having trials:

- ARQ-197

- MetMAb

- XL-184

- Xalkori (crizotinib)

I was under the impression, though, that MetMAb wasn't living up to its early promising results as a means to overcome this mechanism of resistance to EGFR inhibitors; I hope I'm mistaken.  You might find the above in NIH PubMed or ASCO conference abstracts or ongoing clinical trials:

http://www.ncbi.nlm.nih.gov/pubmed

http://meetinglibrary.asco.org/abstracts

http://www.clinicaltrials.gov/

P.S. -- FWIW, I know of patients using crizotinib as an inhibitor of MET amplification (mechanism of resistance erlotinib for EGFR+ lung cancer), one an actual patient who is trying to stay NED, and another written up in a published case in a journal I vaguely recall reading but could be mistaken about. 

I am recently running into this field. I read several review papers and some claim that cMET inhibitor achieve promising Clinical activity with an acceptable toxicity. The other paper said None o f the HGF/cMET inhibitors have shown significant efficacy in clinical trials and no RTK inhibitors primarily targeting cMET have been approved. Which one is correct? I hope experts in cMET field provide some update. Thanks in advance.