Sarah A. Alwardt, PhD

, Marcus Neubauer, MD, chief medical officer for The US Oncology Network, and I kicked off a 2-part series exploring the implications of the disruptions to oncology treatment patterns caused by coronavirus disease 2019 (COVID-19) and the potential impact on patient outcomes.

 looked at how real-world data (RWD) gathered during COVID-19 on treatment patterns, complications, and the viability of treatment modification options can provide insights to inform future cancer care.

Part 2 builds on that foundation, reviewing how COVID-19 also has transformed the normal clinical trial process—trials have been discontinued or delayed, and data collection has been affected. This article will discuss the implications of this disruption and answer 2 critical questions, which are:


how to leverage RWD for clinical trial rescue to maintain drug development during COVID-19, and

how to understand the impact of COVID-19 on the data itself.

Due to efforts to reduce exposure to the coronavirus, oncology patient volumes initially decreased, treatment regimens were disrupted or altered, and new cancer diagnoses also slowed. This affected not only patient care, but also clinical trials.

As a result, many trials were delayed or paused as biopharma sponsors tried balancing patient safety with minimizing risk to trial integrity. Recognizing that trial sponsors faced such challenges as quarantines, hospital and physician 

office closures, travel limitations, and supply chain interruptions, the United States Food and Drug Administration (FDA) issued guidance in March for conducting clinical trials during the COVID-19 pandemic that indicated protocol modifications or deviations might be unavoidable.

Using RWD to Rescue Trials Disrupted by COVID-19 

Yet, even with these challenges, it is critical to maintain the drug development and regulatory process to make new drugs and new treatments available. We have an opportunity to think differently about the trials in flight and revamp the process of gaining insight into treatment patterns and performance during COVID-19 and beyond.

With a goal of protecting patient safety and helping sponsors complete trials and accelerate access to treatments, we must expand the use of RWD to complete trials disrupted due to COVID19. This includes expanding randomized controlled trial (RCT) eligibility and looking at new ways to leverage RWD to provide an alternative to having to accrue 2 arms when trial recruitment is challenging enough without a pandemic. Here are some methodologies to consider:

: Even prior to COVID-19, the FDA had begun to embrace the use of synthetic control cohorts to support drug development and regulatory approval of breakthrough medicines designed to improve patient care. With the willingness of the FDA to accept adjustments to trial protocols, synthetic control arms offer a tremendous opportunity to augment clinical data by allowing biopharma company sponsors to reduce 

the number of subjects assigned to the control arm in a randomized trial, or to conduct smaller randomized trials. This could be critical to completing or restarting trials disrupted by COVID-19.

 According to the National Cancer Institute, standard of care is defined as treatment that experts agree is appropriate, accepted, and widely used. There are a number of affected trials that were using— or planned to use—a prospectively accrued standard-of-care control arm. We have the opportunity to review the protocols for these trials and determine where strong evidence of standard of care exists that would support the use of historical data as the control arm. For example, RWD was pulled from McKesson’s iKnowMed oncology EHR data set to identify patients with metastatic Merkel cell carcinoma (mMCC). Those data were compared with a clinical trial arm for a new therapy to understand the patient response to treatment. This helped establish the FDA approval of a first-line therapy for mMCC, which was the first time RWD was part of a regulatory approval in oncology. This type of trial innovation reduces the number of patients who would need to be accrued during a pandemic, helps maintain patient safety, and avoids potential COVID-19 comorbidity challenges.

Real-world cohort as part of antiviral trials

: In its formal guidance on development of COVID-19 treatments and preventative agents issued in May, the FDA indicated that expansive enrollment criteria will be necessary for clinical 

development to address the scope of the public health crisis and urged sponsors to enroll complicated or fragile patients. The use of pragmatic clinical trials, focusing on the correlation between treatments and outcomes in real-world clinical settings, has increased significantly over the last several years. Pragmatic trials offer the ability to test treatment options on more representative patient populations with fewer exclusions for common conditions that could affect outcomes in practice. These data can be used as a stand-alone analysis or as part of the RCT.

How to Ensure Data Quality During COVID-19

In addition to looking at methods to safely and efficiently conduct clinical trials in this new environment, it is also important to understand the impact of the coronavirus on the data itself, and to ensure that the data are complete, high quality, and generalizable across broad populations.

The potential for COVID-19 infections will reshape how we think about patient data for this time period, especially since testing has not been implemented consistently across the country. With a huge population that may have been affected, how do we think about recruitment, as well as including those we know were infected, 

when we’re analyzing health outcomes during this time frame? Do we simply censor the data and exclude any patients who were diagnosed or treated between January and May 2020, or can we find an appropriate way to address this potential confounding?

In order to track and record information related to the COVID-19 screening population, McKesson developed new data collection tools specific to the coronavirus for its iKnowMed oncology EHR. Beginning in mid-March, practices in The US Oncology Network started to screen patients at the door and gather clinical data about symptoms (fever, cough, fatigue) and risk factors, such as recent local or international travel. Practices also recorded if patients had been tested for COVID-19 and the test results.

To date, practices in The US Oncology Network have screened more than 100,000 patients. This is the beginning of a COVID-19 database that can be used to figure out whether patients’ treatment courses and outcomes were altered, as well as if patients had complications because they were COVID-19 positive, among other things.

With the ability to validate data lineage and verify patients’ status, these data will help identify qualified patients to enroll in trials more quickly.

The ongoing COVID-19 pandemic has magnified the value of RWD in oncology, and the industry has responded with a number of initiatives aimed at refining the role of RWD in guiding clinical trial development, answering important clinical questions, and supporting regulatory decisions. This includes the COVID-19 Evidence Accelerator, a collaborative effort to develop a common set of RWD elements and an initial set of priority questions regarding the therapeutic interventions, risk factors, and the management of COVID-19 in patients with cancer.

McKesson working with the Friends of Cancer Research, the FDA, biopharma companies, and other industry stakeholders on this new paradigm. This data aggregation effort is designed to not only look at how to best care for patients today, but to inform future cancer care and sup-port new models of clinical trials by leveraging RWD to its full potential.

FDA Guidance on Conduct of Clinical Trials of Medical Products during COVID-19 Public Health Emergency. FDA. March 2020. Accessed August 18, 2020. https://bit.ly/3gH9Bab

Sarah A. Alwardt, PhD, is the Vice President of Data, Evidence, and Insights Operations for McKesson

Articles

Due to efforts to reduce exposure to the coronavirus, oncology patient volumes initially decreased, treatment regimens were disrupted or altered, and new cancer diagnoses also slowed. This affected not only patient care, but also clinical trials.









How to Leverage  Real-World Data as COVID-19 Transforms Clinical Trials

Oncology is rapidly transitioning into a field where tumor types are increasingly sub-segmented by molecular characteristics, and treatments are becoming structurally informed and designed based not only on tumor characteristics but also patient characteristics. In the conduct of tumor sequencing, multiple alterations are often discovered. However, alignment between molecular alteration and treatment efficacy is still imperfect. Consequently, it is common to use several drugs aimed at various alterations, rather than just 1. Although this approach may be necessary, it increases the complexity of treatment and the possibility of adverse events. 

Much is still unknown about the cancer process and how targeted therapies work on specific tumor vulnerabilities. Providers must closely follow evolving information, especially regarding adverse events. This knowledge will enable them to be proactive to quickly modify treatment and avoid compromising efficacy while keeping the patient on therapy.

Adverse Effects Differ From Chemotherapy 

Chemotherapy as a treatment modality has been used since the 1940s, so its adverse effects are generally well known. Adverse effects caused by targeted therapies can be different from chemotherapy. Some adverse effects are still under appreciated because, in the context of a clinical trial report, they were not fully described due to low frequency and/or prevalence; however, that does not mean that they are harmless. When a targeted agent combines with other drugs the patient is taking, an unexpected adverse effect may appear and be more severe than those caused by chemotherapy.

Many immunotherapy agents can trigger serious problems, such as autoimmune-based adverse events. Providers must respond quickly to manage these conditions. Some of these adverse effects, such as thrombocytopenia, leukopenia, or anemia, may not require management or can be handled in the office. However, a patient with severe diarrhea or pneumonitis who is taking a checkpoint inhibitor can quickly be in a life-threatening situation that requires hospitalization.

Antibody-drug conjugates (ADCs) are an example of how adverse effects of a parent compound might be 

mitigated while new issues arise. ADCs were developed to provide a “targeted” approach to delivering chemotherapy drugs that might otherwise be too toxic to be systemically delivered in its native form. The conjugates leverage tumor cell expression of antigens and/or receptors to focus delivery of a cytotoxic warhead. For the most part, these conjugates are stable in solution and only release their active agent upon binding and internalization at the cancer cell level. However, unique toxicities may occur due to normal cell expression of the same targets or irritation from the conjugate itself.

Today, there are many different types of targeted drugs, and the knowledge base is still being developed concerning adverse events they may cause. Adverse effects largely depend on the type of drug administered and what it targets. The most common adverse effects associated with certain targeted therapies are diarrhea and liver problems, such as hepatitis and elevated liver enzymes (CHECKLIST).

Clinical Trials Leave Many Questions Unanswered 

Even though new targeted therapies may have received FDA approval, little is known about their performance and safety in the broader patient population. Approval in a clinical trial does not mean the drug is right for everyone. Trial populations are relatively small compared with the prevalent population of patients, and documented adverse events may be similarly challenged–particularly since the manner in which adverse events are reported in the medical literature highlights the most common events (usually ≥10%) and the highest degrees of severity. This underestimates the potential burden of less common and less severe adverse events that could have a major impact on treatment tolerance. Extrapolating that data to the real world leaves many unanswered questions. With so many unknowns, bringing these drugs to a broader population is challenging and requires surveillance–a process called pharmacovigilance.

A major difference in a clinical trial’s study population versus the general at-risk population is how the patients on a trial were selected for participation. To reduce the number of confounding variables that could affect the primary efficacy signal, eligibility criteria are outlined to homogenize the study participants. While eliminating patients with preexisting conditions provides a clearer signal, it also effectively reduces extrapolation value, bringing unknown risks to patients not matching those used for the study. Most drug approvals carry a postmarketing commitment to gather data on real-world evidence, and updated safety packages are provided as new data become available. In some cases, this has led to more restrictive general use because the adverse events were more common than anticipated.

Although drug manufacturers do a good job of compiling data about adverse events, much is still unknown. Simply put, health care providers can find themselves in uncharted waters, trying to navigate the safest way for patients to receive potentially lifesaving therapies with limited information. Providers must understand that targeted therapies can potentially affect normal processes in the body and create adverse effects that may be unknown or under appreciated at the time of study closure. Consequently, providers need to be on alert for all symptoms, even if these symptoms are vague or unattributed, as they may ultimately signal the start of an escalating problem.

Comorbidity is very common among patients with cancer. Most commonly, it’s the rule rather than the exception. Among the top 4 incident cancers–lung, colorectal, breast, and prostate comorbidity rates are 52.9%, 40.7%, 32.2%, and 30.5%, respectively.2 These secondary illnesses 

must be managed to ensure patients are healthy enough to withstand treatment. However, it is well documented that some of these medications can complicate the effective delivery of cancer medicines, particularly drugs where natural metabolism is shared.

Continuing education at the provider level is crucial to the effective and safe management of patients with comorbid conditions. Detailed history of concomitant disease and current medication is essential to developing a comprehensive treatment plan. Increasing complexity of both the oncologic and comorbid condition treatment heightens the need for providers to continually update their understanding of new agents and adverse events, utilizing educational opportunities that focus on these new therapies.

Mitigating strategies can be developed based on data that emerge as therapies become more widely used in patients with comorbidities. This approach has been successful with immunotherapy, especially CAR (chimeric antigen receptor) T-cell therapy. As the severity of the adverse effects began to emerge, toxicity guidelines were developed to help providers manage severe autoimmune adverse events, with special attention paid to patients with comorbidities. Today, providers can receive accreditation for delivering CAR T-cell therapy and other complex therapies. As more advanced treatments appear, accreditation may become a reimbursement requirement.

Enabling Successful Treatment Providers can help ensure optimal outcomes for patients undergoing targeted therapies by employing best practices in several key areas, such as:

Patients and providers should be thoroughly educated about known adverse effects. Pharmaceutical companies have done a good job producing materials that provide pretreatment education concerning adverse effects and when to seek help if experiencing difficulties. With immunotherapies, it is critical for patients and providers to understand when symptoms must be immediately addressed and when they can wait. Certain symptoms can worsen in a matter of hours, creating a life-threatening situation.

Closely monitoring patients enables providers to be proactive about adverse events. Many clinical trials today are actually measuring the time to an adverse event and developing a profile for the drug under study. This is a giant step forward in managing symptoms, as it enables providers to create mitigation strategies and understand the natural cadence of adverse events.

There are several approved targeted drugs that require close monitoring because they can cause severe effects outside of the office between cycles, such as hypertension. When this was recognized during trials, home blood pressure cuffs and logs were provided to patients to track their readings. Utilizing this information, a combination therapy was identified that created dangerous adverse effects within the first 2 weeks, enabling providers to proactively manage these symptoms. When the drugs were finally approved, these best practices were brought to the office level to mitigate adverse effects.

Certain PARP inhibitors also require closer monitoring, particularly for platelets and bone marrow suppression. Consequently, weekly counts after the first treatment cycle are required with some of these agents to identify potential problems.

Many practices in The US Oncology Network are using triage and symptom tracking via patient portals to assess degrees of urgency. This real-time communication helps ensure patients get immediate care if needed, minimizing adverse events.

Staying Current Is Challenging But Critical 

The rapid pace of emerging targeted therapies is creating a dynamic world that is increasingly complex for providers. Keeping up with the huge volume of information required to offer these therapies is difficult to manage in addition to patient care responsibilities. Undoubtedly, this will be an ongoing challenge.

The key to success may lie in using all forms of instructive technology and presentations focused on targeted therapies, including social media, webinars, publications, seminars, and other offerings. These educational opportunities enable providers to share real-world evidence and experiences, helping to build a critical knowledge base. As more new agents emerge, every available avenue should be utilized to keep the nuances of these complex therapies top-of-mind for providers so they can deliver safe, expert care.

 1. Targeted cancer therapies. National Cancer Institute. Accessed September 14, 2020. https://bit.ly/2ZQohyl 

2. Pirschel C. The impact of comorbidities on patient care. ONS Voice. January 30, 2017. Accessed September 14, 2020. https://bit.ly/35Ahe0v

Much is still unknown about the cancer process and how targeted therapies work on specific tumor vulnerabilities. Providers must closely follow evolving information, especially regarding adverse events.

Optimizing Benefit in Targeted Therapies Requires Multipronged Educational Approach

disease 2019 (COVID-19) pandemic and stay-at-home orders swept across the country in March, oncology care—like nearly every aspect of the US health care system—was dramatically affected. As efforts to reduce exposure to the coronavirus ramped up, nonemergency surgical procedures were halted, screening procedures were delayed, and the amount of new cancer diagnoses declined. In addition, some clinical trials were shelved, and treatment regimens were disrupted due to COVID-19 illness or fear of getting infected. In a 2-part series, we will explore the implications of these disruptions to oncology treatment patterns and the potential impact on patient outcomes. In this article, we will look at how real-world data (RWD) on treatment patterns, complications, and the viability of treatment modification options gathered during COVID-19 can provide insights to inform future cancer care.

Unprecedented Impact on Cancer Care Oncologists normally are well-equipped to make clinical decisions regarding treatment for their patients. However, COVID-19 has added an unprecedented layer of complexity, forcing oncologists to predict what would happen if surgery, chemotherapy, or radiation were delayed due to state restrictions or COVID-19 hardships. Questions to ponder include: Will treatment increase the risk of coronavirus infection? Will delaying treatment allow for the growth or spread of the cancer? Are there alternative treatments that are less risky but perhaps not as effective? What if patients lose their jobs and health insurance?

COVID-19 Enhancements to Electronic Health Records

To get a longitudinal view of the patient journey, it is critical to evaluate clinical facts, such as relevant biomarkers, disease staging, and disease progression. This information, along with key demographic and patient characteristics, is widely available through real-time tracking of clinical and claims data elements across unif ied electronic health records (EHR) and reimbursement systems. McKesson’s iKnowMedSM oncology EHR captures outpatient medical histories from community oncology practices in The US Oncology Network and has been used to help understand the real-world utilization and outcomes associated with a number of oncology agents.

In addition to looking at the impact of treatment disruptions, it is also important to ensure the data are complete, high quality, and generalizable across broad populations. In order to track and record information related to COVID-19, McKesson developed a new feature in its EHR that lets practices in The US Oncology Network document results of screening patients at the door.

Beginning in mid-March, practices started gathering clinical data about symptoms (fever, cough, fatigue) and risk factors, such as recent local or international travel. Practices also are recording if patients were tested for COVID-19 and the test results. To date, practices in The US Oncology Network have screened more than 100,000 patients. This is the beginning of a COVID-19 database that can be used to determine whether patients’ treatment courses and outcomes were altered, as well as if patients had complications because they were positive for COVID-19.

Analyzing RWD and the COVID-19 database will enable the oncology community to innovate for the future by building predictive models and algorithms that help us inform cancer care decisions. While we are just beginning to analyze the data, there are some interesting early indicators: 

: We’re starting to see a variety of treatment modifications, such as dose reductions and switching to schedules with a longer interval between treatments when indicated. We expect there will be a lot to learn about what happens when doses are modified or time is extended between treatments—information that can be used for updated labeling.

: Due to stay-at-home orders and hesitation to travel long distances, patients who might have normally gone to an academic medical center or larger hospital are seeking second opinions and treatment at community oncology clinics.

: Telemedicine offered immediate relief to patients—and caregivers— wanting to reduce their exposure to the coronavirus. Although every visit cannot be virtual, this represents a huge shift in oncology care.

During the stay-at-home process, most cancer surgeries and biopsies were considered elective. One analysis that we plan to conduct is to look at the impact of delays in diagnosing and staging disease, and, ultimately, starting treatment. Specifically, we want to determine whether more patients are presenting with more advanced disease. If we see more initial diagnoses of advanced disease than we saw before the pandemic, we have data to support changing practices so that biopsies are not elective in future events.

Outcomes and changes to treatment patterns associated with treatment delays, longer intervals, or delays in diagnoses and staging cannot be tested in a standard randomized clinical trial. However, as we’ve seen during the pandemic, these data are critical to helping clinicians make informed decisions on behalf of their patients. We are proud to work with the oncology community to leverage RWD to learn from this pandemic and provide critical insights to inform future cancer care.

Marcus Neubauer, MD, is the Chief Medical Officer for The US Oncology Network, and Sarah A. Alwardt, PhD, is the Vice President of Data, Evidence, and Insights Operations for McKesson

As efforts to reduce exposure to the coronavirus ramped up, nonemergency surgical procedures were halted, screening procedures were delayed, and the amount of new cancer diagnoses declined.

Sarah A. Alwardt, PhD | Authors

Articles

How to Leverage Real-World Data as COVID-19 Transforms Clinical Trials

Optimizing Benefit in Targeted Therapies Requires Multipronged Educational Approach

Using Real-World Data to Understand the Impact of COVID-19 on Oncology Patient Care