If the pursuit of precision medicine were a race, molecular data might seem like the sprinter who burst out of the starting gate. Molecular data have fueled groundbreaking advancements, powering new therapies and diagnostics that have accelerated our fight against some of humanity’s most challenging diseases. Consider what happened during the COVID-19 pandemic: over 5 million viral genomes were sequenced globally to track variants, a clear demonstration of the tremendous potential of molecular insights (1).
Yet, despite serving as the foundation of diagnosis for over 150 years, pathology data is now shifting into high gear. It’s not just keeping pace – it’s rapidly moving into a leadership position in this new era of care.
On your mark: Understanding molecular and pathology data
Before we zoom to pathology’s growing impact, let’s review how we got here. Molecular data – including DNA and genomics – offer a unique window into what makes each of us biologically distinct, making them a natural catalyst for breakthroughs in personalized medicine. Molecular data are also highly quantifiable, reinforcing their value in research and diagnostics.
Though molecular data provide crucial insights into the individual, they yield only an indirect understanding of the disease he or she may face. These data reveal a portion of the blueprint that contributes to the condition but lack direct phenotypic representation of the condition itself.
Pathology, on the other hand, offers a clear picture at the tissue, cellular, and subcellular levels. For decades, though, its promise has been limited by traditional microscopy, depending largely on qualitative assessment and the human eye’s interpretive ability. As a result, extracting deeper insights about disease has been somewhat elusive.
Until now.
Get set: Whole slide images are laying the foundation for pathology’s precision medicine era
Quality and efficiency gains might be the commonly cited advantages of digital pathology, but another significant benefit deserves recognition; each whole slide image captures billions of pixels providing one of the most direct – and detailed – profiles of diseases like cancer. These images introduce a new data modality that’s roughly ten times richer than a typical radiology image.
Moreover, whole slide images make it possible to apply AI to pathology data – a breakthrough that represents the biggest development in the field since the introduction of the light microscope 150 years ago. AI taps into the wealth of data within these images – unlocking insights invisible to the human eye and quantifying characteristics that could reshape our understanding of disease.
Go! Unleashing the power of pathology data
Among their many use cases today, AI algorithms count mitoses in breast cancer biopsies, determine the prevalence of biomarkers, and provide objective data for tumor grading. These tools bring to pathology the same objectivity that gave molecular data its early edge.
Compared to their molecular counterparts, pathology-based tests are often cheaper and faster to run so that more patients can start the best treatment sooner. They also typically preserve more of the original tissue since DNA, RNA, or proteins do not need to be extracted – making it easier to conduct further testing down the line. So, does this mean that pathology data is poised to pull ahead of molecular data?
Not so fast.
A team effort, not a race
Though we are clearly entering a new era of pathology, it isn’t one that will be defined by a competition among data types. Instead, we are at the start of realizing the impact of an increasingly collaborative, multi-modal approach.
Precision medicine, at its core, hinges on both the individual and the disease. Molecular and pathology data form a dynamic duo that work in concert – likely even with other modalities – to provide understanding unlike anything before. They are not racing against each other but rather accelerating towards a shared finish line: better health outcomes for all.
Newsletters
Receive the latest pathology news, personalities, education, and career development – weekly to your inbox.
References
- G Berno et al., “SARS-CoV-2 variants identification: overview of molecular existing methods,” Pathogens, 11, 9 (2022). PMID: 3614590.
