Harvard Tissue Atlas Harvard Medical School
Data

High-plex immunofluorescence imaging and traditional histology of the same tissue section for discovering image-based biomarkers

Lin JR, Chen YA, Campton D, Cooper J, Coy S, Yapp C, Tefft JB, McCarty E, Ligon KL, Rodig SJ, Reese S, George T, Santagata S, Sorger PK

Nature Cancer (2023). DOI: 10.1038/s43018-023-00576-1

Precision medicine is critically dependent on better methods for diagnosing and staging disease and predicting drug response. Histopathology using Hematoxylin and Eosin (H&E) stained tissue - not genomics – remains the primary diagnostic method in cancer. Recently developed highly-multiplexed tissue imaging methods promise to enhance research studies and clinical practice with precise, spatially-resolved, single-cell data. Here we describe the “Orion” platform for collecting H&E and high-plex immunofluorescence images from the same cells in a whole-slide format suitable for diagnosis. Using a retrospective cohort of 74 colorectal cancer resections, we show that IF and H&E images provide human experts and machine learning algorithms with complementary information that can be used to generate interpretable, multiplexed image-based models predictive of progression-free survival. Combining models of immune infiltration and tumor-intrinsic features achieves a nearly 20-fold discrimination between rapid and slow (or no) progression, demonstrating the ability of multi-modal tissue imaging to generate high-performance biomarkers.

Publication | Dataset | Colorectal Cancer Atlas | Response article from Nature Cancer

Key Findings:

  • The Orion platform rapidly collects 18-plex immunofluorescence (IF) and diagnostic-grade H&E images from the same cells

  • Complementing H&E with IF (i.e., multimodal imaging) provides greater insight into the molecular features of colorectal cancer samples

  • Machine learning models trained on the multimodal and clinical data predict progression-free survival

  • Analyzing multimodal imaging using machine learning / artificial intelligence methods reveals biomarkers not discoverable by other means

Click to enlarge.

Top: Overview of the Orion method for same-slide immunofluorescence and H&E imaging. Bottom: Example images of a colorectal cancer specimen from Orion, showing the multiplex immunofluorescence image (left; 5 of 18 fluorescence channels shown), the corresponding same-section H&E image (center), and the overlay of these two images (right).

Contents

Data Explorations

Data image
Multimodal spatial profiling of colorectal cancer using Orion

Data Overviews

These Data Overviews provide access to minimally processed Level 2 images with no annotation or quality control. Click any of the following thumbnail images for an interactive view of the full-resolution images.

Data image
Orion CRC01 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC04 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC07 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC10 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC12 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC13 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC14 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC15 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC16 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC17 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC18 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC19 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC20 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC21 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC22 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC23 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC24 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC30 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC31 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC33_02 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC35 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC36 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC37 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC38 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC39 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC40 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC02 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC03 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC05 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC06 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC09 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC11 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC08 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC25 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC26 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC27 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC28 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC29 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC32 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC33_01 - overview - Lin, Chen, Campton et al., 2023
Data image
Orion CRC34 - overview - Lin, Chen, Campton et al., 2023

Data access

The images in Lin, Chen, Campton et al. (2023) comprise a ~10.5 TB dataset. Visit doi.org/10.5281/zenodo.7637655 to view a table detailing where to find each file and data type. This data will eventually also be available via the NCI-sponsored repository for Human Tumor Atlas Network (HTAN; https://humantumoratlas.org).