The single-agent activity of MEK inhibitors in MAPK or CDK4/6 inhibitors in cyclin pathway aberrant tumors has been limited. The combination of trametinib and palbociclib demonstrates safety, tolerability, and clinical activity in a histology-independent manner, representing a therapeutic approach for patients harboring co-occurring aberrations.
See related article by Kato et al., p. 2792
The primary objective of the cabinet game “whack-a-mole” is to use a mallet to hit as many moles as possible when they pop up in random succession. The challenge here is, as you hit one mole, two or three more pop up and some even disappear before you can hit them all making it impossible to hit all the moles. Whack-a-mole game is a perfect metaphor for our attempt at targeted therapies for precision oncology. While monotherapy leads to initial success, resistance often develops, warranting combination therapies (Fig. 1).
In this issue of Clinical Cancer Research, Kato and colleagues (1) successfully used a novel combination of commercially available CDK4/6 and MEK inhibitors to target cooccurring alterations in cell-cycle and MAPK pathways. Prior clinical data have shown that targeting either of these pathways with the same agents alone is in-effective (2). Equally important was their inclusion of notoriously difficult-to-treat histologies, such as pancreatic adenocarcinoma (6/9 patients in this study). For these multiple therapy–resistant patients (median two prior lines), a clinical benefit rate of 56% was reported with minimal toxicity. The authors demonstrated several important principles of targeted therapy and precision oncology:
(i) Drugs can be used to inhibit a pathway rather than a specific oncogenic driver/mutation.
(ii) Drugs do not have to be given at their full recommended or approved dose to inhibit a pathway when used in combination to balance toxicity, safety, and efficacy.
(iii) Drugs that block downstream of an oncogenic driver need assistance from alternate pathway blockade.
(iv) Combination therapies can overcome or delay resistance.
(v) Combination therapy safety and toxicity for customized combinations can be managed at the N of 1 level.
(vi) Multi-disciplinary molecular tumor boards are needed to move beyond one-gene-one-drug conventional clinical trial paradigm to address the reality that each tumor is unique and complex at the molecular level (3).
The authors' work builds on a rich history of targeted therapy starting from chronic myelogenous leukemia (BCR-ABL) in heme malignancies and gastrointestinal stromal tumors (c-KIT) in solid tumors, to more recently non–small cell lung cancer (EGFR, ALK, ROS1, BRAF, and RET). These diseases represent the first step of personalized medicine: inhibition of a constitutively activated strong oncogene driven by a point mutation, insertion/deletion, or gene fusion. The second step, represented by BRAF/MEK inhibitor combination in BRAF V600–mutated melanoma, is hallmarked by dual blockade of both a strong oncogenic driver and another element in the same pathway. We now enter step three: customized combination therapy inhibiting two different pathways without direct inhibition of an oncogene.
Previously, the proof of principle came from a similar strategy involving combination of an aromatase inhibitor with a CDK4/6 inhibitor in hormone receptor–positive breast cancer. The estrogen receptor signals through a complex pathway that eventually leads to cyclin D activation and cell-cycle progression (4). The dual inhibition of these pathways has had dramatic impact on the survival and quality of life of patients with breast cancer. The current study builds on and reinforces pathway inhibition as a therapeutic modality by including and indeed highlighting patients with currently undruggable KRAS non-G12C mutations. The previous failure of both drugs as single agents in RAS aberrant tumors implies that the tumor easily bypasses a block by invoking an alternate pathway, but has a more difficult time subverting blockade at divergent points. In this case, blockade of the dominant pathway (likely MAPK) requires assistance from far downstream cyclin inhibition.
The main clinical challenge with combination therapies is that synergistic or additive efficacy may also lead to synergistic or additive toxicity. This study offers some insights into this issue in the way it was conducted, not as a phase I dose escalation study, but as a molecularly matched observational study. The drugs were procured off-label by medication acquisition specialists and not by predetermined planned arrangement with sponsors. The investigators were free to use doses other than what was recommended in combination (5), tailor to tolerance, escalating or deescalating, and titrating as dictated by the individual patient not bound by the confines of a clinical protocol. This study benefited from prior combination of these agents done in a phase I trial (5), which established two different recommended combination regimen (RCR) doses, both higher than what was used by Kato and colleagues. It would be impossible to guess this dosing without such prior data. The authors chose a lower than RCR dose, capitalizing on synergy between the agents. Synergy may lead to synergistic toxicity and even drugs, such as trametinib and palbociclib, which have few overlapping expected toxicities, are intolerable when combined at their full doses. Undoubtedly, there will be many other opportunities for multi-pathway blockade and the question of safety in dosing will continue to arise. With the rise of molecular tumors boards, the question is especially apropos. Recommendations for such combinations will be made, but dose guidance has to accompany these propositions. Do we reject such novel combinations until a phase I study is done? Do we accept a blanket “rule of thumb,” such as using half the recommended dose for each drug? This seems overly simplistic. Or do we create an N of one dose escalation phase I, where the drugs are started at low doses and titrated to tolerance and possibly efficacy? Starting too low may not benefit the patient. Starting too high may be risking intolerable toxicity leading to discontinuation. These challenges are compounded when we move beyond two-drug combinations.
An outstanding question is whether aberrations of CDKN2A are necessary for this strategy to succeed. Perhaps blockade of the cyclin complexes in combination with blockade of any upstream activated pathway is sufficient to achieve synergy with or without an aberration in the cyclin genes (6)? This can be answered in a subsequent trial of trametinib and palbociclib in KRAS-mutant cancers with or without CDKN2A/B aberrations.
Finally, this study represents the beginning of a new paradigm in drug development for target therapies. Perhaps more important than successful co-inhibition of MEK and CDK4/6 is the precedent this investigation sets. First, this study will encourage molecular tumor boards at institutions with precision oncology capabilities to recommend similar combinations. Admittedly, translating this to real-world practice may be challenging, especially for monitoring unknown or additive toxicities Second, this validates the broad-scale use of next-generation sequencing in patients with cancer. Third, the inhibition of pathways rather than individual genes/mutations greatly increases the scope of targeted therapy and this study lights a path forward with such an approach. Fourth, the clear signal in pancreatic cancer warrants a larger trial using the MEK/cyclin inhibitor combination. Fifth, molecularly matched therapies may be moved up earlier in the treatment plan before other resistance mutations arise, saving mutation burden–dependent immunotherapies for later. Sixth, as precision oncology expands to include combination therapies, this study is clearly a preview to the next iteration of NCI-Match. Whereas the first iteration mostly tested single drugs, the upcoming NCI-comboMATCH will test combinations of targeted drugs to overcome drug resistance to single-agent therapy. Finally, to win a “whack-a-mole” game of targets in precision oncology, we may need more than one mallet and several players from a multi-disciplinary molecular tumor board to hit hard, hit fast, and hit them all.
V. Subbiah reports grants from LOXO Oncology during the conduct of the study and Bayer outside the submitted work, and research funding/grant support for clinical trials (to his institution) from AbbVie, Agensys, Alfa-sigma, Altum, Amgen, Bayer, Berghealth, Blueprint, Boston Biomedical, Boston Pharmaceuticals, D3, Dragonfly Therapeutics, Exelixis, Fujifilm, Idera Pharma, Incyte, InhibRx, Loxo oncology, Medimmune, MultiVir, NCI Cancer Therapy Evaluation Program, National Comprehensive Cancer Network, Novartis, PharmaMar, Pfizer, Takeda, Turning Point Therapeutics, and University of Texas MD Anderson Cancer Center. No disclosures were reported by the other author.
The authors thank the inspiration from Dr. Emil J. Freireich, MD (1927–2021), father of modern oncology clinical trials who first pioneered combination trials in oncology. V. Subbiah acknowledges support by NIH grant R01CA242845. UT MD Anderson Cancer Center, Clinical Center for Targeted therapy was supported by the Sheikh Khalifa Bin Zayed Al Nahyan Institute for Personalized Cancer Therapy, 1U01 CA180964, NCATS grant UL1 TR000371 (Center for Clinical and Translational Sciences), and the MD Anderson Cancer Center Support Grant (P30 CA016672).