# Preclinical Tumor Growth Delay Is More Reliable from Imaging-Based Rather than Manual Caliper Volume Measurements

## TL;DR

This study shows that imaging-based tumor volume measurements are more reliable than manual caliper measurements for assessing treatment effects in preclinical cancer research.

## Contribution

The paper demonstrates that MRI-based tumor growth delay estimates are more accurate than caliper measurements in preclinical oncology studies.

## Key findings

- Caliper measurements showed poor correlation with MRI-derived tumor growth delay rates.
- Volume differences between caliper and MRI measurements increased as tumors grew.
- Only 15% of caliper-derived NTGD rates were within 10% of MRI-derived rates.

## Abstract

Background/Objectives: Tumor growth delay is frequently used in preclinical experiments evaluating oncologic interventions. While treatment response in humans is based on imaging criteria for obvious reasons, manual caliper measurement of subcutaneous tumors is standard in animal studies. In a murine tumor model treated with immunotherapy (ImT) and radiotherapy (RT), the reliability of caliper measurements was tested by comparing normalized tumor growth delay (NTGD) rates derived from caliper- and image-based volumetrics. Methods: A 4T1 breast syngeneic murine model was used, in which thirty animals were inoculated in the right inguinal mammary fat pad and the right axilla. One RT fraction of 8 Gy was delivered to the right inguinal tumor on day 11 post-implant, and intraperitoneal ImT (PD-1 checkpoint inhibitor) injections were administered on days 11, 12, and 14. Each animal underwent three MRI scans (days 10, 17, and 20). Caliper measurements were also performed by two independent observers on the same days. The measurements were averaged and used to estimate ellipsoid tumor volumes. The acquired MRIs were used for image segmentation and volume estimation. Tumor volumes (days 17 and 20) were normalized against the baseline pre-treatment tumor volume (day 10). NTGD rates derived from hand- and image-based volumetrics were compared to assess the reliability of caliper vs. MRI estimation. Results: Caliper volumes between the two observers correlated at 0.799 (Pearson, p < 0.001). The averaged caliper volumes correlated with MRI volumes at 0.897 (Pearson, p < 0.001). Absolute volume differences between caliper and MRI increased with tumor growth. NTGD-derived rates showed no correlation, with only 15% of NTGD caliper rates falling within 10% of the MRI rates. Conclusions: NTGD rate based on caliper volumes is a suitable measure of treatment response in preclinical studies. In the experiment described herein, caliper-derived NTGD rates did not correlate with MRI ground truth. These findings suggest that more accurate tumor volumetrics, derived from stored and verifiable medical imaging sources, should be used in preclinical assessment of oncologic interventions instead of standard caliper estimates.