Spatial transcriptomics and current standard care approaches represent fundamentally different paradigms for analyzing tumor biology and informing clinical decision-making. While spatial transcriptomics provides unprecedented molecular resolution within tissue architecture, it remains inaccessible to the vast majority of cancer patients due to significant cost, complexity, and infrastructure barriers. This comparison examines the technical capabilities, clinical applications, practical limitations, and emerging bridging technologies between these two approaches.
Current standard cancer care relies primarily on histopathological examination using H&E (hematoxylin and eosin) staining, a century-old technique that provides morphological assessment of tumor architecture, cellular organization, and tissue composition 1). H&E analysis enables pathologists to identify malignant cells, assess differentiation status, evaluate immune infiltration, and determine tumor margins. The approach is rapid, cost-effective (typically $50-200 per sample), and universally available across clinical laboratories worldwide.
In contrast, spatial transcriptomics maps gene expression across intact tissue samples while preserving spatial context 2). Technologies such as 10x Visium, MERFISH, and seqFISH capture RNA molecules at micrometer-scale resolution, generating thousands to millions of spatial data points per sample. This molecular mapping reveals cell-type composition, gene regulatory networks, microenvironmental gradients, and tumor-immune interactions with extraordinary detail—making it “the richest way to read a tumor” according to recent assessments. However, spatial transcriptomics assays cost $3,000-$15,000 per sample, require specialized equipment, demand bioinformatic expertise, and typically extend analysis timelines by weeks to months.
The accessibility gap between these approaches is dramatic. Approximately 0% of cancer patients in standard care currently receive spatial transcriptomics analysis, despite its scientific superiority, due to cost-prohibitive pricing, limited laboratory capacity, and lack of established clinical workflows. The global spatial transcriptomics infrastructure—measured in dozens of specialized research centers—cannot scale to serve the millions of cancer patients diagnosed annually. Standard H&E histopathology, by contrast, is performed on virtually every cancer tissue sample as routine diagnostic care across thousands of laboratories in every country.
Current standard care primarily supplements H&E with immunohistochemistry (IHC) for specific biomarkers (HER2, PD-L1, ER/PR), next-generation sequencing (NGS) for somatic mutations and targetable alterations, and occasionally flow cytometry for hematologic malignancies. These modalities provide clinically actionable information but lack the spatial context that transcriptomics preserves, and they typically assess only 5-500 genes rather than the 18,000-20,000 genes profiled by spatial platforms.
Recent developments have focused on predicting spatial transcriptomic maps from routine H&E images using deep learning-based prediction models. TARIO-2, a computational framework announced in 2026, represents a significant breakthrough in this direction 3). This approach leverages the observation that H&E morphology contains substantial information about underlying molecular composition, cellular identity, and gene expression patterns. By training neural networks on paired H&E images and spatial transcriptomics data from research cohorts, TARIO-2 can generate predicted spatial gene expression maps from H&E slides that patients have already undergone.
This bridging strategy offers several advantages: it eliminates the cost differential (utilizing existing H&E samples), accelerates time-to-result (converting existing tissue samples instantly), requires no additional laboratory resources, and maintains compatibility with established clinical workflows. The predicted maps may enable clinicians to extract some spatial molecular insights without requiring patients to undergo expensive spatial transcriptomics assays.
Standard care advantages include universal accessibility, immediate results, established clinical interpretation guidelines, regulatory approval for decision-making, and proven prognostic value across decades of clinical practice. Its primary limitations are restricted molecular scope (morphology and limited biomarkers only) and loss of spatial information in tissue homogenates.
Spatial transcriptomics advantages include comprehensive gene expression coverage, preserved spatial architecture, discovery of novel biomarkers, and potential identification of clinically relevant microenvironmental features (immune infiltration patterns, metabolic gradients, vascular proximity effects). Limitations include prohibitive cost, limited patient access, non-standard clinical interpretation, lack of validated decision algorithms, and extended turnaround times.
For the foreseeable future, spatial transcriptomics will remain a research and precision medicine tool for selected patients rather than a routine standard-of-care modality. Clinical decision-making in oncology will continue to rely on H&E histopathology supplemented by targeted biomarker testing and genomic sequencing. The value proposition of bridging technologies like TARIO-2 lies in gradually democratizing access to spatial molecular information by leveraging infrastructure already embedded in clinical workflows, potentially enabling broader scientific insights while standard-of-care pathways remain centered on proven, accessible modalities.