Thermal Imaging as a Predictive Biomarker to Assess Clinical Response in Carcinoma Cervix Treated with Radical Radiation Therapy

Abstract

<h3>Purpose/Objective(s)</h3> Cervical cancer is one of the leading malignancies in women, and concurrent chemo-radiation followed by brachytherapy is the current standard of care for locally advanced carcinoma cervix. There are various biomarkers including serum, molecular and imaging biomarkers that can predict the response to chemo-radiation therapy. This study evaluated the role of Thermal imaging as an imaging biomarker in predicting the response to chemoradiotherapy by measuring the variation in tumor surface temperature between responders and non-responders during treatment, end of treatment and on follow-up. <h3>Materials/Methods</h3> A single institution, prospective observational study was conducted from April 2020 until August 2021. After receiving informed consent, all patients with biopsy-proven locally advanced Carcinoma Cervix who were planned for concurrent chemoradiation, and brachytherapy were recruited. Thermal images from the tumor surface were taken at baseline, 2 weeks after initiation of EBRT (interim), at the completion of EBRT (end of RT), and on first follow up (6-12 weeks after completion of treatment). The change in temperature during treatment (ΔT1), end of treatment (ΔT2) and on first follow-up (ΔT3) were correlated with clinical and radiological responses during and post-treatment. <h3>Results</h3> The mean age of patients was 52 years with a range of 34 - 74 years, and the commonest stage at presentation was FIGO stage IIIC1. Nodal involvement was seen in 55% and involvement of parametrium was seen in 95% patients. Baseline mean temperature value was 39.33°C (SD±0.63), with a range of 38.10 to 40.20°C. Temperature values were measured on all the thermal images, and it was found to decrease during the treatment. The sensitivity and specificity of ΔT to predict response was calculated using ROC curves and a cut off value of temperature was obtained using Youden index to differentiate responders and non-responders. End of treatment ΔT2 > 1.65°C correlated significantly (p=0.047) with response to radiation therapy, with a sensitivity of 87.5% and specificity of 50%. First follow-up ΔT3 > 4.15°C correlated significantly (p=0.023) with response to radiation therapy, with a sensitivity of 87.5% and specificity of 75%. <h3>Conclusion</h3> A serial decrease in temperature is observed during treatment and on follow-up. This decrease is significantly higher in responders as compared to non-responders. Thermal Imaging is a feasible non-invasive tool that can be considered as an adjunct to clinical assessment to predict treatment responses in cervical cancer patients treated with definitive radiation therapy. The change in temperature (ΔT) at end of treatment and on first follow-up can identify potential non-responders and to optimize treatment.