We recently interviewed Mark Kaganovich, CEO of Tracer Biotechnologies, a precision medicine company working on decentralizing MRD testing. See the full transcript below.

‍

Hi Mark, thank you for joining us. To start, could you please share a bit about your background and about Tracer Biotechnologies?

After finishing my BA in Biochemistry and Computer Science and earning a PhD in Genetics, I co-founded a software company called SolveBio, which focused on helping pharma R&D teams organize and automate data collection in clinical trials. The company was later acquired by Precision Medicine Group, and soon after, Gopal and I started Tracer. Our experiences at SolveBio gave us perspective on how innovation in pharma works, and we came away convinced that some of the most meaningful opportunities in life sciences are those that accelerate the feedback loop between clinical trials, patient treatment, and diagnostics. Noninvasive quantitative cancer detection is a strong example of this, and one where software expertise can make a significant difference.

At Tracer, our work centers on developing sensitive, precise, and cost-conscious approaches to solid tumor cancer detection. Our goal is to maximize sensitivity, accuracy, and scalability while keeping the technology practical to implement. This pursuit has led us into two key technological areas: digital PCR (dPCR) and whole genome sequencing (WGS), each with unique strengths for different aspects of cancer detection.

Our dPCR product takes advantage of the technology’s capacity to detect single nucleotide variants with high precision and efficiency, particularly when biomarker information about a tumor is already available. In these cases, we have created processes to design and validate personalized dPCR assays to monitor circulating tumor DNA in a patient’s bloodstream.

Our WGS approach is designed for situations where tissue is not available or higher sensitivity is required. Using a proprietary algorithm, this assay analyzes high-throughput sequencing reads to detect the presence of circulating tumor DNA without requiring complex, custom molecular biology.

These two approaches may seem different, but what unites them is the intent: to make cancer detection more sensitive, reliable, and accessible. Over time, we believe that such technologies will become routine, not limited to specialized labs, but widely available across healthcare systems. For that to happen, factors like ease-of-use, turnaround time, and decentralization will be critical. One way we move toward that goal is by transferring much of the complexity into software, which is more adaptable and scalable across environments.

‍

While MRD has been advancing rapidly, it is still a relatively early space, and I have seen many different descriptions of MRD or variations of its full form. Do you consider "MRD" to stand for "Minimal Residual Disease" or do you consider a different full form to be more appropriate for the abbreviation?

When we say MRD, we mean “Minimal Residual Disease”, although, to be precise, we are measuring molecular markers of disease that are a proxy for residual disease. There is still a lot of work to be done to connect molecular (or imaging) detection with clinical relevance. Eventually all of these things will be consolidated into one term that indicates whether the patient has a pathology or not and what one can do about it.

We honestly prefer to specifically talk about ctDNA detection. MRD is usually used in the context of recurrence monitoring, which is a very important application. Another critical application is monitoring patient response to treatment. Sparing patients unnecessary treatment may end up being as impactful an application as detecting pathology early.

‍

What do you see as the main applications for MRD?

ctDNA-based disease quantification, including MRD, is having a transformative impact on clinical care and drug development. There are two areas where the impact is significant: (1) detecting pathology significantly earlier than existing methods, e.g. disease recurrence or ultimately in the screening setting and (2) precise disease burden quantification which will flood the industry with more data on longitudinal dynamics of disease as a response to therapeutic regimens.

For the first part, data on the clinical actionability and benefits of early detection of recurrence and accurate detection of ctDNA clearance is starting to trickle in, so both clinicians and drug developers are adopting these tools. This in turn impacts clinical development. If you can confidently say “there is no disease” earlier than other methods, or if you know that intervening earlier will have better outcomes, that changes and likely accelerates clinical development and commercialization of therapeutics.

On the second part, measuring longitudinal disease dynamics precisely has real implications on multiple levels. For clinical care, giving physicians the ability to accurately measure if a treatment is working in as close to real time as possible is transformative. This can lead to adjusting treatment or de-escalating. Of course, there are a lot of questions and there will need to be many studies to truly connect dynamics with clinical benefit. Collecting this kind of data, especially piggy backing on early indications of utility, will answer a lot of these questions and likely be a universal component of clinical care.

‍

Are there any barriers to MRD adoption that you see? And if so, any thoughts on how to overcome them?

I wouldn’t phrase it in terms of barriers, but rather accelerants – or lack thereof. When there is clear evidence of clinical action and benefit as a result of detecting ctDNA, utilization will take off. Clinicians and patients will drive payer coverage and clinical adoption. There is a positive feedback loop of evidence driving accessibility driving adoption, but there’s always a chicken and egg dynamic in the industry, and that usually means a steeper adoption curve in the early days.

‍

What are some trends you see in the liquid biopsy and MRD space overall? Anything you forecast or predict in the near future?

Our big prediction at Tracer is that there will be a convergence of several trends that result in new paradigms in the industry. For example:

(1)   ctDNA accepted as either a regulatory disease readout for trial outcome and/or commercially critical data for drug companies to generate as part of their effort to demonstrate why their drug is better than alternatives.

(2)   Underlying technologies improving in efficacy and cost: e.g. DNA sequencing perhaps plateaued for a while and we think it is now (back to) accelerating. Similarly, throughput of dPCR instrumentation and the technology and workflows around that platform have been improving.

(3)   Decentralization: there will be many more locations that want to run these kinds of tests. Rather than one massive lab, customers (whether that is pharma, CROs, or hospital systems) will internalize these capabilities for a variety of reasons: lowering cost, accelerating turnaround time, maintaining competency, and/or collecting a data asset.

‍