Optimal Scheduling of First-Line Therapeutics in Non-Small Cell Lung Cancer

In the United States, lung cancer causes more deaths per year (135,720 estimated for 2020) than does any other type of cancer, and 84% of lung cancer deaths are caused by non-small cell lung cancer (NSCLC) (Siegel et al., 2020). Treatment for NSCLC has expanded greatly in the last century. Early in the 20th century, the primary mode of treatment for NSCLC was surgery, but today we have a large range of therapeutic options available to patients (Jászai & Schmidt, 2019; Liu et al., 2017). An important advancement in oncology was the pairing of chemotherapy drugs (chemical agents which directly attack tumors) and targeted therapy (drugs which precisely target singular mechanisms of tumor proliferation). Between 1996 and 2010, it's estimated that the percentage of NSCLC patients receiving some combination of targeted therapy and chemotherapy has increased by 20% for stage I/II patients, 28% for stage IIIA patients, and at least 15% for stage IIIB/IV patients (Kaniski et al., 2017). Parallel with advancements in treatments, the prognosis for these patients has also been improving. The 5-year survival rate in NSCLC has advanced from 10.7% in 1973 to slightly less than 21% as of 2019 (Lu et al., 2019). Today, several combinations of chemotherapy, immune checkpoint inhibitors (drugs which prevent cancer cells from evading the immune systems), and antiangiogenics (therapeutics designed to reduce blood flow to tumors) are recommended as first-line therapies for the management of metastatic or recurrent NSCLC (Lung Cancer - Non-Small Cell - Types of Treatment, 2012).

Some patients are burdened with NSCLC that is not affected by chemotherapy. Generally, this phenomenon is called treatment resistance, and in the case of chemotherapy it is called chemotherapy resistance. Chemotherapy resistance that is present at diagnosis is called intrinsic resistance, where chemotherapy resistance that develops over the course of the disease is called acquired resistance. Chemotherapy resistance (intrinsic or acquired) is a regular occurrence in NSCLC. In a 2006 study of resected NSCLC, Thomas A. d’Amato et al. measured significant chemotherapy resistance to carboplatin in 68% of samples, to cisplatin in 63% of samples, and to paclitaxel in 40% of samples – all first-line therapeutics for NSCLC (d’Amato et al., 2006). In the KEYNOTE-001 trial for pembrolizumab (a clinical trial which established the clinical efficacy of pembrolizumab in NSCLC) only 19.4% of patients responded to treatment. The low response rate indicates that most patients did not significantly respond to treatment (Garon et al., 2015). Over a long enough period of treatment, almost all NSCLC becomes broadly treatment resistant (Chang, 2011).

Theoretically, dosages could simply be increased to improve efficacy in cases of treatment resistant NSCLC. However, chemotherapeutics, antiproliferatives, and immune checkpoint inbhibitors have narrow therapeutic windows. This means that the range between a dosage large enough to reduce tumor growth, and small enough to be safe, is small. In cases of acquired or intrinsic resistance to chemotherapy, dosages cannot be easily increased without producing serious adverse effects, including gastrointestinal disturbances, immunosuppression, and anemia (Ahmad & Gadgeel, 2016). Due to both the high side-effect burden and high rate of resistance, NSCLC patients are typically moved to second-line or experimental therapies over the course of their treatment. The majority of NSCLC clinical patients are using drugs still in development (Non-Small Cell Lung Cancer Treatment (PDQ®)–Patient Version - National Cancer Institute, 2020). There is, therefore, a critical need to improve efficacy of both first-line therapeutics and experimental therapeutics to improve therapeutic outcomes, patient survival, and ultimately patient quality of life.

Mathematical modeling and computer simulation of pharmacology (pharmacometric modeling) is an extremely efficient method for optimizing therapeutic efficacy. Pharmacometric modeling does not require the considerable time and resource investment required to conduct a suite of clinical trials in humans. To develop the model, the pharmacometrician can leverage data from multiple studies, involving diverse patient populations, and varying drug combinations and administration schedules, to build a complete mathematical description of the therapeutics and disease. After building the model, it can be used to simulate a series of “what if?” scenarios (e.g. what if the dose was cut in half, but given twice as often?), and to derive the best scheduling and dosing of various therapeutic drug interventions based on the characteristics of the patient.

We have previously published a mathematical model of NSCLC growth dynamics which we used to demonstrate that administering bevacizumab (BEV) and pemetrexed/cisplatin (PEM/CIS) sequentially with a gap of 1 day, rather than concurrently, would improve the efficacy of this combination (quantified as final tumor volume) by more than 50% without the need for increasing therapeutic doses. However, optimal scheduling of this combination in humans has yet to be verified with large sets of clinical data, and the model needs to be extended to explain the behavior of other therapies, including immune checkpoint inhibitors.

Our proposed research consists of two primary aims. First, we expect to use a large set of clinical data to establish a mathematical model to describe tumor growth and response to various drug combinations currently in use for NSCLC (Aim 1). In addition, we expect to determine individual patient characteristics that significantly affect treatment response. This mathematical model will be used to determine the optimal dosing schedule of therapeutic interventions in individual patients with NSCLC (Aim 2). This contribution will be significant as it will broadly and positively impact current healthcare outcomes and clinical therapeutic development in a highly prevalent and largely intractable disease.