The pharmaceutical industry continues to make great strides to provide new medicines to patients. Over the course of the last 10-20 years, oncology therapeutic advancement has proceeded stepwise from chemotherapy, to targeted therapies, to improvements on successful targeted therapies, and into truly patient-specific treatment. We’ve entered an amazing new era – the official arrival of personalized medicine. Our previous article discussed the hype of Chimeric Antigen Receptor T-Cells (CAR-T) as a personalized medicine – but as a view from the finish line. Here we look at the development path towards a personalized medicine to understand the hurdles.
Background – Cancer Pathogenesis
Cancer is an accumulation of various capabilities (e.g. mutations or cellular changes) that results in the ability of the cancer to grow unchecked, evade immune surveillance, and several other key characteristics. These “hallmarks” are defined and then updated by Drs Hannahan and Weinberg in a series of sentinel works that summarize the body of literature on cancer growth and evasion. Basic science research into the mechanisms of cancer growth are the bedrock on which drug development is built. This research elucidates the inner workings of a cancer cell and generates targets that could be drugged to treat the tumor. Once a target is identified, a drug can be designed (or found) to hit that target and can proceed through non-clinical and clinical testing.
Understanding these hallmarks of cancer enabled the growth of targeted and personalized therapy. For instance, many cancers have aberrant signaling pathways that contribute to, or are essential to the growth of, the tumor. One such signal is associated with growth factor signaling, and the tyrosine kinases associated with these growth factor receptors. About 538 tyrosine kinases have been found by analyzing the human genome, and the U.S. FDA has approved 35 tyrosine kinase inhibitors. Yet other therapies seek to exploit overexpression of proteins on the surface of tumor cells – such as the external domains of these same growth factor receptors, or immune markers.
Why do some patients respond to treatment and others do not? The era of ‘precision,’ ‘targeted,’ and ‘personalized’ medicine arose from the understanding of tumor mutations causing different patients to have different results with the same medicine. Some of the approved labels restrict use to patients with a given mutation or protein expression; other treatments are not restricted. Shepherding a drug from target to optimization to clinic to approval is a skilled artform where expertise is key at every turn.
Research & Development
Part of the discovery process is a decision around how a druggable target will be actioned. Can you use a molecule to inhibit the signaling from a receptor, or do you need to use a biologic to block the receptor extracellularly? Certain tumors have known mutations of signaling receptors, leading to differential efficacy of an inhibitor in those patients. These approaches could interfere with tumor cell growth and starve the cell. Or does the overall expression of this receptor on the cell surface enable targeted approaches with a biologic or CAR-T? The CAR-T would bind the receptor and kill nearly every cell expressing these proteins.
These topics are key to the understanding of the mechanism of action of most treatments, and though there are exceptions, these should be data-driven decisions led by a team of experts. As experiments proceed through in vitro and in vivo testing, and data are generated, drug development experts create and hone the eventual label to be achieved with the treatment (the Target Product Profile). In addition to creating scientific rationale for the treatment, the team must learn to manufacture the product, and in the case of personalized treatments such as CAR-T, this is a key divergence in complexity.
We’ve seen certain mutations identified a priori allowing targeted therapy clinical development in a selected patient population (i.e. only enrolling patients expressing that mutation, or enrolling additional patients expressing that mutation). Yet other mutations are discovered after clinical trials have begun. Finding a select population where a treatment has differential efficacy can maximize the effect measure, thereby minimizing the number of patients needing to get to market. Erlotinib is a good example of a targeted therapy in this context: the drug is approved in certain indications for all patients, but there is differential efficacy for those patients with mutations in the EGF Receptor. Further research has revealed more about the ways in which tumors have become resistant to EGFR inhibitors which has enabled development of new drugs to emerge with better efficacy (e.g. osimertinib, brigatinib, alectinib).
The concept behind patient-specific treatments is to use this knowledge about drug targets and apply advanced techniques to increase tumor cell death in the process. Most of these advanced techniques use a viral payload to add or remove genes from a patient’s own cells. There is an elegant complexity to these treatments. First, these products are not manufactured in bulk as every dose is different. The patient’s cells are removed, re-engineered, and re-administered to the patient on-demand, placing the manufacturing timeline into the patient’s treatment scheduling and timeline. This adds a layer of uncertainly to clinical development as the efficacy is tested alongside the manufacturing; teasing the two elements apart can be challenging. Did the treatment fail because of a manufacturing issue, or because the treatment truly doesn’t work?
Oncology trials always warrant careful oversight from sponsor and CRO teams, but in these situations a dosing error – giving one patient’s treatment to another patient – could trigger a lethal immune reaction. Expertise in overseeing these trials is paramount to the continued success of the treatment. PRA has managed many trials in this space, and leverages lessons learned to continually improve our oversight model and team structures in a Center of Excellence model.
Refinement of the clinical development plan proceeds as data are collected: efficacy data, toxicity data, biomarker data, PK/PD data (as appropriate), but also data on the manufactured product. Though it’s difficult to generalize the types of changes that may be warranted, there is no substitute for experience so that no signals are missed. These products have shown a unique set of toxicity that requires diligence from the investigator and study team. For instance, lethal cases of cytokine release syndrome have been reported which require swift intervention.
In parallel to the cell therapy wave, we’re also watching gene therapy for treatment of monogeneic rare diseases. These therapies are similarly patient-specific with the aim of curing certain rare diseases where a single gene is driving the disease state. Replacing this one gene throughout the body can potentially cure the disease.
Drug development is a scientific endeavor with a healthy dose of creativity. Never has this been more true with the era of patient-specific treatments. Scientific progress brought these first therapies to market, and successive generations of these therapies can now iterate on the science to improve efficacy and safety. Expertise in development oversight is key to unraveling the treatment, manufacturing considerations, and the path(s) to get these treatments to patients.
Brandon Early is a drug developer with deep oncology and immunology experience. In his current role as VP, Clinical Development at PRA Health Sciences, he serves as Chief of Staff within the core CRO business, Product Registration. Prior to PRA, Brandon ran global clinical development programs in the biotech and CRO sector. He received his Masters and Undergraduate degrees from the University of Virginia.