Harnessing lab data to speed time to treatment

In recent years, pharma companies have begun to embrace lab data as a richly rewarding dataset for a whole host of purposes, including trial enrollment and site identification, patient monitoring and disease progression, and commercial targeting and market access insights. Along with test orders, values, and results, deidentified lab data includes a wealth of information, such as a patient’s demographic information, diagnosis codes, and ordering provider and facility details.

In addition to providing a deep understanding of a specific patient’s clinical and biochemistry profile, lab data is often the key driver in diagnostic and therapeutic decisions. It’s also timely, which is an important distinction. Unlike retrospective claims data, lab data is almost immediately accessible to manufacturers, making it particularly valuable to pharmaceutical companies that want to reach patients in time to enroll them in clinical trials or impact their treatment plan.

Diverse utility for every therapeutic area

As lab data is so close to the patient, it provides pharma companies with valuable insights into how care progresses at a patient level. The sheer volume of lab tests ordered daily makes this dataset instrumental to pharma companies tracking the patient journey. For example, our partner Quest Diagnostics generates more than 500 million new lab test results per year for 3,500+ lab tests, including more than 900 genetic tests. As Quest’s collection also spans 15 years of longitudinal data, that’s a significant and ever-growing dataset.

While pharma companies in every therapeutic area derive significant value from lab data, oncology manufacturers are currently the most frequent users. In oncology, pharma companies use biomarker signaling to understand which patients are indicated for a particular therapy, with use cases ranging from protocol development to market access concerns. Cardiology is also a prevalent therapeutic area.

In recent months, more manufacturers in the neurology space have begun leveraging lab data, most notably for identifying Alzheimer’s and Parkinson’s patients. Manufacturers of rare and frequently misdiagnosed conditions also use lab data to analyze current clinical workups and track which factors contributed to previous misdiagnoses.

Although some pharma companies turn to lab data for a limited use case, the most sophisticated continually leverage data across the product life cycle. For example, one manufacturer began working with Quest Diagnostics years ago to identify disease prevalence for an underdiagnosed condition. After the company decided to develop a new therapy, it used lab data to recruit patients for early-stage trials and then to identify investigators and determine optimal trial sites. Once the therapy was approved, the company then used lab data for commercial targeting purposes to drive provider outreach.

 Harmonized data reveals the patient’s journey

To provide the most utility for pharma companies, real-world datasets must first be linked and tokenized at the patient level. Integrated, harmonized real-world data allows manufacturers to examine the entire patient care journey to uncover and solve patient access barriers. At Norstella, we can combine real-world claims, lab, and EMR data with our proprietary forecasting, clinical, regulatory, payer, and commercial intelligence data to create a fully integrated data asset.

We build our clinical profiles and deliverables based on a client’s specific needs, whether for clinical targeting, research, or commercial targeting. We typically start with the lab data from the local and reference labs for a comprehensive view and build from there, adding closed and open claims data.

We also incorporate structured and unstructured EMR data, such as the clinical notes written by the physician when a patient comes in for a clinical workup. The EMR data allows us to review an in-depth clinical history of the patient, including testing and imaging results, scoring systems, and the physician sentiment around clinical decision-making to build our profiles on qualitative and quantitative data insights.

While lab data remains the anchoring dataset for Norstella’s clinical profiles, adding the patient’s procedure history, comorbidities, diagnoses, and treatments enables endless data analysis possibilities. Together, this integrated data allows us to map a patient’s complete clinical journey. As all our data sources are tokenized on a patient ID, we do not know the patient’s identity—but we can quickly identify the patient’s treating physicians and monitor their progress with our linked data assets. Our NorstellaLinQ helps us fill in the gaps associated with typical real-world data assets, enabling a 360-degree view of the patient journey.

Myriad use cases across the product life cycle

So, how can lab data be used across the product life cycle? Let’s look at a few use cases.

Early detection of eligible patients is always challenging, but for rare and ultra-rare disease clients, patient identification is especially difficult—particularly in the development phase. One client of Quest Diagnostics, for example, could only identify a single patient for its clinical trial in an eight-month period via traditional methods. After using Quest’s deidentified lab data, they could enroll six additional patients within two months. This represents the difference between a viable clinical trial and one that would need to be halted, so the results were highly significant for the client.

Many of Norstella’s clients have recently used lab data analysis to impact patient treatment directly. One example was a commercial targeting project in the multiple myeloma space. We built profiles both for newly diagnosed patients and for patients with relapsed or refractory multiple myeloma (RRMM) who needed to switch therapies as their current treatment was no longer working. Across these patient profiles, we identified more than 800 new patients for our client. We used lab-based alerts for specific test results to trigger proactive provider outreach to those ordering NPIs, which resulted in many of these patients starting on the client’s therapy.

Due to the longitudinal nature of lab data, pharma companies can gain unique insights into disease progression. Which patients’ levels are changing over time? This kind of data is immensely useful for patient cohort definitions and for surfacing NPIs for ordering providers.

Sometimes, lab data can uncover new opportunities for a drug that has been on the market for a while. For example, one manufacturer of a hepatitis C therapy linked census and income data to Quest lab data to identify zip codes with lower-than-expected testing rates given the population. Using this evidence, the pharma company contacted providers in those regions to educate them about the testing disparities, leading to increased diagnoses and prescriptions.

Relying on data science and clinical expertise 

This last use case highlights the importance of collaboration, as the Hepatitis C data analysis initially emerged from a dialogue about disease prevalence. Any successful data analysis begins with asking the right questions and finding the right experts to guide you through that process.

At Norstella, we have extraordinary in-house clinical expertise to help clients determine what they should be querying, which clinical time points are involved, and how to interpret the resulting insights. Our strong team of clinicians and data science experts collaborate to harmonize thresholds and naming conventions to make this data usable and ingestible for our clients.

Scoping a new project involves detailed conversations about the results a client is seeking and the types of labs involved. Does a client want a test ordered by itself or within a broader panel? We help educate our clients about the nuances of lab testing because it opens up potential use cases they might have yet to anticipate. For example, one manufacturer determined that a new, more advanced sequencing test was impacting provider ordering behavior, which led to fewer diagnoses—an unexpected but actionable insight for the client.

Deidentified lab data offers immense value to pharma companies. Proactive companies are using this robust data to accelerate time to treatment, whether in the clinical setting or for existing therapies.

To learn more, watch my discussion with Patrick Winniewicz, executive director of Healthcare Analytics Solutions at Quest Diagnostics, in this on-demand Fierce Pharma webinar: Level Up Your Lab Data: New Use Cases for Improving Time to Treatment.