Answers are grounded in this paper and cite their sources.
Improved understanding of cancer biology fueled by a rapid escalation in our ability to harness multidimensional population data, coupled with diversification in therapeutic modalities through innovations in biotechnology, have transformed oncology therapeutics in recent years. This issue of Clinical Pharmacology and Therapeutics (Figure 1) provides a tour of selected innovations in precision medicine, model-informed drug discovery and development frameworks, and the evolving clinical trial and regulatory landscape that have nurtured patient-focused oncology research, development, and clinical practice. Drug discovery, development, and clinical practice in oncology have witnessed a sea change in recent years. Therapeutic hypotheses linked to actionable molecular alterations are enabling precision medicine with both molecularly targeted agents and immunotherapies across diverse cancer types. Rational dose selection is being increasingly practiced in drug development through the application of quantitative pharmacology methods, including the rapidly growing discipline of systems medicine that bridges pharmacokinetic-pharmacodynamic approaches and molecular pharmacological sciences to optimize the development and use of precision medicines. Innovations in trial design and regulatory frameworks, increasingly adopting a totality of evidence Bayesian learning mindset have enabled acceleration of drug development, allowing timely access to innovative therapies for patients. Long-term durable disease control is becoming possible with current and emerging therapies, at least in some patients and tumor types. Therefore, a patient-focused approach to cancer treatment will need to not only focus on efficacy but importantly on quality of life, making it increasingly important to minimize the gap between drug development and real-world clinical practice. For much of the history of oncology, treatment approaches have been based on a “one size fits all” approach, with the recognition that some individuals will have a significant response to therapy, some may have a more limited response, and others will not have any response to treatment. Over time, as the field’s understanding of the key pathophysiological mechanisms underlying the behavior of specific cancer types evolved, so too did the treatment regimens. However, the problem still remains that responses to therapy are heterogeneous and difficult to predict. With the more recent explosion of high-throughput assays that enable “omics,” including genomics, epigenomics, proteomics, and metabolomics, it is now possible to obtain a wealth of information about a specific patient’s tumor. This has led to the burgeoning field of precision cancer therapeutics, in which an analysis of a patient’s tumor determines the therapeutic approach. It is hypothesized that this precision-medicine approach will be more effective than the traditional paradigm of picking a therapy that has proven to be effective in a heterogeneous patient population.1 Some malignancy types, such as acute myeloid leukemia, have only recently begun to move from the “one size fits all” approach of cytotoxic chemotherapy to using cytogenetic and somatic mutations to guide therapy choices.2 The field of solid-tumor oncology has begun to develop tissue agnostic therapies, such as the use of immune checkpoint inhibitors for microsatellite instability-high, or mismatch repair-deficient tumors, or the use of agents that target neurotrophic tropomyosin-related kinase fusions.3, 4 Tissue agnostic therapies, as well as precision oncology in general, require the identification of biomarkers and the development of uniformed testing for these biomarkers.5 However, thus far, there continues to be heterogeneity in the assays used to measure particular biomarkers, uncertainty as to how to define the cutoffs that will determine whether a therapy will be offered to a particular patient, and uncertainty as to how to incorporate the results of many different biomarkers in the case of a complex system, such as the immune response.5 Furthermore, deep sequencing analysis of intratumoral and subclonal heterogeneity is revealing new insight into tumor complexity and resistance mechanisms.6 Consequently, it is becoming increasingly evident that a “one size fits all” approach will not be effective, even if a precision-based approach was used, if that therapy does not target all of the clones and resistance mechanisms. Thus, much work remains to be able to optimally match therapies to a patient’s specific tumor.7 In addition, it is clear that a precision oncology approach will also need to incorporate a personalized medicine approach to supporting therapies8 and an improved understanding of how commonly used medications impact tumor biology.9 Quantitative clinical pharmacology considerations play a central role in optimizing dosing of oncology therapies to maximize benefit vs. risk.10 For example, young overweight patients appear to be at increased risk of subtherapeutic endoxifen levels highlighting a potential opportunity for individualized dosing in this subpopulation.11 In the case of the kinase inhibitor regorafenib, in patients with colorectal cancer, higher unbound exposures of two active metabolites were associated with poorer clinical outcomes related to early treatment discontinuations, suggesting an opportunity for dosage individualization.12 Whereas therapeutic drug monitoring (TDM) appears promising for many anticancer agents, actionable TDM targets are often lacking. Pediatric TDM targets for targeted kinase inhibitors have been recently proposed, applying principles of extrapolation.13 Although primarily in a population with nonmalignant disease receiving treosulfan-based conditioning prior to allogeneic hematopoietic stem cell transplant, estimation of nonmonotonic exposure-response relationships for complex outcomes (mortality-free engraftment) enabled definition of an optimal exposure range for pediatric TDM.14 Whereas these examples are illustrative of hypothesis generation for TDM, equally crucial for clinical translation are innovations in model-informed clinical decision support systems for deployment of TDM and downstream patient management in clinical practice.15 Physiologically-based pharmacokinetic modeling and simulation has steadily driven efficiency in evaluating intrinsic and extrinsic factor effects in drug development to inform dosing and administration across clinical contexts of use, as elegantly demonstrated with the cyclin-dependent kinase 4/6 inhibitor ribociclib.16 Of note, our understanding of sources of variability in the clinical pharmacology of therapeutics continues to evolve, illustrated in a survey of the impact of hepatic impairment on pharmacokinetics of monoclonal antibodies.17 It is evident that the practice of clinical pharmacology in the era of novel therapeutic modalities (like gene and cell therapies, oncolytic viruses, and personalized cancer vaccines) will warrant unique considerations and the development of new approaches.18 The blurring of the lines between learning and confirming in contemporary cancer drug development calls for a totality of evidence mindset, leveraging all available preclinical and clinical data to optimize selection of patient population, dose, dosing schedule, combination, and trial design. Mechanism-informed quantitative data integration is indispensable in the development of innovative biotherapeutics in immuno-oncology.19, 20 With increasing biological complexity of the underlying mechanisms of action, the question at hand is one of multidimensional optimization (dose, schedule, combination, and patient population), necessitating application of holistic mechanism-based quantitative translational frameworks.21, 22 It will be impossible to evaluate all possible combination partners and sequences of administration in all tumor immunophenotypes. Quantitative systems pharmacology (QSP) models permit simulations of expectations in virtual patient populations that recapitulate the heterogeneity of the tumor microenvironment, enabling rational pursuit of precision medicine, and combination hypotheses in clinical development. Investigational immunotherapies, such as bispecific antibodies, have the potential for bell-shaped dose-response relationships and potential for cytokine release, making dose selection for first-in-human trials a nontrivial undertaking and emphasizing the central importance of QSP frameworks.23 Systems models can revolutionize precision medicine by enabling adaptive personalization of treatment decisions for individual patients through iterative validation of “personal models” using biomarker and response data collected dynamically during individual patient journeys.24 Given the explosive increase in the generation of “big data” across multiple dimensions (-omics, imaging, and clinical outcomes) in the development and use of oncology therapeutics, convergence and synergy across mechanistically driven (e.g., QSP), and biologically agnostic (e.g., statistical approaches leveraging artificial intelligence/machine learning) models will be key to success.25 This will require partnerships between quantitative pharmacologists, data scientists, cancer biologists, and clinicians in an increasingly integrated cross-sector ecosystem that nurtures fundamental biological inquiry in the context of patient-focused real-world practice. The increasing momentum toward precision cancer therapeutics has led to new challenges in oncology clinical trial design and regulatory frameworks. The traditional phase II–III trials that evaluate efficacy of a novel therapy in a defined patient population (e.g., adjuvant therapy for localized colorectal cancer) are difficult to conduct if the therapy in question is tissue agnostic (and, therefore, encompassing many different malignancy types) or involves rare mutations or rare cancers. Novel clinical trial designs, including basket trials, umbrella trials, and platform trials, are increasingly being performed to more rapidly evaluate novel therapies that are matched to specific tumor characteristics.1, 26, 27 In addition, small single-arm trials, which have primarily used overall response rates as the primary end point, have been used to obtain conditional regulatory approval.28 There are many challenges associated with how subsequent confirmatory trials with survival end points can be conducted in these settings.26 Furthermore, with the rapid development of novel therapies, there is increasing pressure for regulatory agencies to evaluate these novel therapies in an expeditious manner. As a consequence, regulatory agencies, such as the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) are developing new evaluation tools and procedures to streamline their regulatory processes and increase access to clinical trials for patients.29-31 In addition, the partnering of regulatory agencies with other key stakeholders early in the drug development process and clinical trial design process will be critical for the advancement of precision cancer therapeutics.29-32 As our understanding of cancer biology deepens and treatments become increasingly tailored to the molecular pathophysiology of a patient’s tumor, it is imperative that we bridge the gap between patients enrolled in clinical trials and the real-world setting. Only 8% of patients diagnosed with a new cancer enroll in clinical trials, in part due to restrictive eligibility criteria, emphasizing a need for rational broadening of eligibility criteria in oncology clinical trials applying the foundational principles of clinical pharmacology as applied to benefit/risk assessment.31 Now more than ever while we are amidst the coronavirus disease 2019 (COVID-19) pandemic, it is imperative that we adopt a patient-focused approach both in clinical research and clinical practice to enable progress in oncology drug development and patient care, leveraging advances in telemedicine and other adaptations that are transforming the broader clinical research ecosystem.33 Furthermore, clinical pharmacology considerations for dose selection are vital in guiding appropriate utilization of drugs used in cancer medicine that may be repurposed for management of COVID-19.34 Envisioned as a component of the optimal patient-focused use of cancer therapeutics, the emerging discipline of interventional pharmacoeconomics utilizes clinical pharmacology principles to develop evidence for off-label treatment regimens that maintain equivalent therapeutic efficacy with reduced costs for patients and health systems.35 Although anticancer drug development has traditionally focused on maximizing efficacy, the safety of cancer treatments and quality of life considerations for patients undergoing cancer therapy has gained increasing attention in the past decade. Advances in molecular pharmacology and real-world data analytics are enabling discoveries of intrinsic and extrinsic sources of variability in dose-limiting toxicities of cancer therapies, as demonstrated elegantly through iterative reverse and forward translational analyses for the microtubule-targeting agents.36, 37Given the complexity of the oncology healthcare ecosystem, critical to real-world evidence generation is the availability of high-efficiency and high-fidelity data extraction from electronic health records, enabled by advances in data science.38 Continued progress in real-world evidence research and its prospective integration into drug development will be particularly vital to close the gap between clinical development and practice and to accelerate access to value-added innovative cancer medicines worldwide.
No takes yet. Share an insight, caveat, or question.
Venkatakrishnan et al. (2020) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: