Few-Shot Prototype Adaptation of Frozen Pathology Foundation Models for Cross-Tissue and Cross-Cancer Transfer: From Oral Squamous Cell Carcinoma to Breast Cancer Lymph Node Metastases

Background/Objectives: Deep networks trained to recognise oral squamous cell carcinoma (OSCC) in histopathology perform well on held-out images from the same collections, but a recent study found that such a network lost most of its discriminative ability on lymph node tissue. We examined whether encoder fine-tuning or label-free adaptation changes this, and how many labelled target patches are needed to recover performance. Methods: The source domain comprised 23,290 images of oral and oropharyngeal mucosa from three public collections. The target domain was PatchCamelyon, a public dataset of sentinel lymph node patches from breast cancer patients; its metastases are breast carcinoma, so the transfer crosses both tissue type and cancer histology. We used 40,000 target patches from 216 slides, split into slide-disjoint adaptation and evaluation pools. We compared nine encoders in ten configurations, including three pathology foundation models (Phikon, Phikon-v2 and the recent Midnight model). We evaluated direct transfer with frozen and fine-tuned encoders, label-free adaptation, and few-shot adaptation, in which we compute class prototypes from k labelled target patches using a frozen encoder. All results are at patch level. Results: Direct transfer failed for every configuration (external AUROC 0.35–0.64), including fine-tuned networks (mean over three seeds 0.44–0.52) and fine-tuned Phikon (0.54–0.58). Label-free methods did not close the gap; a zero-shot vision–language model reached 0.77. With a frozen Phikon encoder, prototypes from 10 and 50 labelled patches per class reached 0.869 ± 0.056 and 0.930 ± 0.015, against 0.82–0.86 at 50 patches for encoders without histology pre-training and 0.977 for a classifier trained on 20,000 labelled patches. Logistic regression on the same patches performed similarly at one and two patches per class and better from five upward (0.955 at 50). Midnight reached 0.922 with prototypes at 50 patches per class, similar to Phikon. The results were similar when support and evaluation patches came from different scanning centres. Conclusions: For the methods and datasets evaluated, no classifier trained on oral tissue transferred to this cross-tissue, cross-cancer target. Our analyses suggest that the change in what the classes contain mattered more than stain or scale, but they do not separate the causal contributions of tissue, cancer histology and acquisition. Frozen pathology foundation models made label-efficient classification on the target domain inexpensive; this replaces the oral-trained classifier and does not use oral images. The findings concern patch-level classification of breast cancer metastases and do not establish clinical performance for OSCC lymph node metastases.

Authors

Institutions

Publication Details

Journal
Cancers
Published
2026-11-12
DOI
https://doi.org/10.3390/cancers18223624
Primary Topic
AI in cancer detection
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Few-Shot Prototype Adaptation of Frozen Pathology Foundation Models for Cross-Tissue and Cross-Cancer Transfer: From Oral Squamous Cell Carcinoma to Breast Cancer Lymph Node Metastases

Mohammad Hassan Ali Al-Onaizan, Mohammed Saleh Ali Muthanna, Samia Allaoua Chelloug, Rashadul Islam Sumon et al.
Cancers
AI in cancer detection
article

Few-Shot Prototype Adaptation of Frozen Pathology Foundation Models for Cross-Tissue and Cross-Cancer Transfer: From Oral Squamous Cell Carcinoma to Breast Cancer Lymph Node Metastases

Mohammad Hassan Ali Al-Onaizan, Mohammed Saleh Ali Muthanna, Samia Allaoua Chelloug, Rashadul Islam Sumon, Salma Akter, Abdul Quddus Mozumder
article en

Abstract

Background/Objectives: Deep networks trained to recognise oral squamous cell carcinoma (OSCC) in histopathology perform well on held-out images from the same collections, but a recent study found that such a network lost most of its discriminative ability on lymph node tissue. We examined whether encoder fine-tuning or label-free adaptation changes this, and how many labelled target patches are needed to recover performance. Methods: The source domain comprised 23,290 images of oral and oropharyngeal mucosa from three public collections. The target domain was PatchCamelyon, a public dataset of sentinel lymph node patches from breast cancer patients; its metastases are breast carcinoma, so the transfer crosses both tissue type and cancer histology. We used 40,000 target patches from 216 slides, split into slide-disjoint adaptation and evaluation pools. We compared nine encoders in ten configurations, including three pathology foundation models (Phikon, Phikon-v2 and the recent Midnight model). We evaluated direct transfer with frozen and fine-tuned encoders, label-free adaptation, and few-shot adaptation, in which we compute class prototypes from k labelled target patches using a frozen encoder. All results are at patch level. Results: Direct transfer failed for every configuration (external AUROC 0.35–0.64), including fine-tuned networks (mean over three seeds 0.44–0.52) and fine-tuned Phikon (0.54–0.58). Label-free methods did not close the gap; a zero-shot vision–language model reached 0.77. With a frozen Phikon encoder, prototypes from 10 and 50 labelled patches per class reached 0.869 ± 0.056 and 0.930 ± 0.015, against 0.82–0.86 at 50 patches for encoders without histology pre-training and 0.977 for a classifier trained on 20,000 labelled patches. Logistic regression on the same patches performed similarly at one and two patches per class and better from five upward (0.955 at 50). Midnight reached 0.922 with prototypes at 50 patches per class, similar to Phikon. The results were similar when support and evaluation patches came from different scanning centres. Conclusions: For the methods and datasets evaluated, no classifier trained on oral tissue transferred to this cross-tissue, cross-cancer target. Our analyses suggest that the change in what the classes contain mattered more than stain or scale, but they do not separate the causal contributions of tissue, cancer histology and acquisition. Frozen pathology foundation models made label-efficient classification on the target domain inexpensive; this replaces the oral-trained classifier and does not use oral images. The findings concern patch-level classification of breast cancer metastases and do not establish clinical performance for OSCC lymph node metastases.

CancersVol. 18(22)
Inje University (KR), Princess Nourah bint Abdulrahman University (SA), Sohar University (OM), Middle East University (JO), Stanford University (US)
Princess Nourah Bint Abdulrahman University
Good health and well-being
AI in cancer detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon