Histotyping

One of the products emerging from the DoMore! project is called “Histotyping”, which uses the marker DoMore-v1-CRC, and culminated in 2020 in a landmark artificial intelligence (AI) publication in the Lancet (Skrede et al., The Lancet 2020).

Histotyping involves an automated analysis of high-resolution scans of cancerous tissue stained with routine hematoxylin and eosin (H&E). ICGI developed convolutional neural networks for this method and trained on extensive retrospective clinical trial data with known patient outcomes. Deep learning models were trained to identify and analyse the morphology of a tumour in a surgically resected tissue section. Histotyping estimates the probability of a good or poor outcome, i.e., the likelihood of a recurrence. It could be employed as a supplementary diagnostic service to support clinicians and patients in their adjuvant chemotherapy decision-making for patients
with stages II and III CRC.

Illustration Histotyping — Histotyping as explained for the Norwegian TV channel TV2.

The DoMore-v1-CRC marker has also been combined with established clinical pathology markers (T and N stage, and number of harvested lymph nodes) in a decision tree solution for clinical decision support. Compared to conventional risk stratification for adjuvant therapy, the proposed method identified a significantly larger group of patients with a good prognosis that are candidates to be spared adjuvant treatment (Kleppe et al., Lancet Oncology 2022)

While an increasing number of publications on deep learning systems claim performance comparable with or better than clinicians, few have yet demonstrated real-world medical utility. We published a Perspective in Nature Reviews Cancer in 2021 to discuss the reasons for the moderate progress and to present our recommendations for how to perform pivotal validations of deep learning systems that will provide realistic performance estimates of the systems and thereby better guide further research and clinical implementation of truly promising systems (Kleppe et al., Nat Rev Cancer 2021).

Using the DoMore-v1 network, a similar approach as for colorectal cancer has been applied to prostate cancer. The adaption of this method from CRC to PCa included using sections from multiple tissue blocks as input images, as prostate tumours are far more heterogeneous than colorectal tumours. The resulting marker, the DoMore-v1-PCa marker, predicts the risk of recurrence after radical prostatectomy and provides prognostic information supplementary to the clinical and pathological markers currently used in clinical practice. A manuscript reporting on the pivotal validation of this marker has been submitted for publication.