Authors : Dr. Nrithi.S.Prasad; Dr. Ramkumar G; Dr. Prasanth Penumadu; Dr. Biswajit Dubashi

Volume/Issue : Volume 8 - 2023, Issue 12 - December

Google Scholar : http://tinyurl.com/28emnwwt

Scribd : http://tinyurl.com/2tx3sdnn

DOI : https://doi.org/10.5281/zenodo.10431851

Abstract : Introduction Oral cancer is a grave problem in many parts of the world. One third of oral cavity cancers occur in India and it accounted for approximately 30% of all cancers in India. Assessment of the change in tumour burden is an important aspect of the clinical evaluation of cancer treatment. Both tumour shrinkage and disease progression are important endpoints in cancer clinical trials using chemotherapeutic drugs. The standard approach used to assess the response of solid tumors to neoadjuvant chemotherapy is RECIST. How ever it has been a common observation that RECIST which uses unidimensional measurement in imaging may not be accurate in response evaluation in oral cavity cancers owing to its complex anatomy. Hence this study was undertaken to evaluate the accuracy of RECIST and to determine if CT volumetric assessment and three dimensional assessment can better evaluate the response. There are only few studies from India in assessing response of oral cavity cancers to neoadjuvant chemotherapy. More over there is a need for more studies to be done so as to find out exactly which criteria is the better one.Objective: To compare the response assessment in CT scan by RECIST, 3D and volumetric methods in borderline operable oral cancers who are receiving neoadjuvant chemotherapy using clinicopathological criteria as the reference standard.Materials & Methods All patients with oral cavity cancers who received neoadjuvant chemotherapy in oncology department of JIPMER and whose CT images were available in PACS were included in the study based on the inclusion and exclusion criteria. Response assessment was done only for the primary tumor by comparing the post neoadjuvant chemotherapy imaging with pretreatment imaging by three different methods namely RECIST, three dimensional assessment and volumetric assessment. In RECIST criteria, single longest measurement in the axial plane was taken. In 3 Dimensional assessment, longest diameters in three orthogonal plane in multiplanar reformatted CT image was taken. In volumetric assessment, manual delineation of tumor extent in serial 2mm axial section was done for computed volume estimation Subtraction of volumes measured at pretreatment and post treatment gave the volume changes. Percentage of reduction in the tumor measurements using the criteria were assessed. Clinicopathological downstaging/upstaging was used as the reference criteria in order to compare the other 3 criteria. The patients were categorized into progressive disease, stable disease, partial response and complete response based on the percentage response. Progressive disease and stable disease were grouped as non responders and partial response and complete response as responders The statistical analyses were performed using SPSS 19 and Medcalc Software. Sensitivity, specificity, positive predictive value and negative predictive value were calculated for all three methods using clinico-pathological assessment as reference standard. Weighted kappa was used to assess the agreement between all three methods for categorizing the patients into progressive disease, stable disease, partial response and complete response. Bland Altman plot was used to compare the quantitative response based on CT volumetry, 3 D assessment and RECIST. To make the volumetric data in 3 dimensional assessment and CT volumetry comparable to unidimensional data of RECIST, cube root of the volumetric data was taken.Results Totally 28 patients with oral cavity carcinoma undergoing neoadjuvant chemotherapy were studied. The sensitivity of RECIST criteria to differentiate between responders and nonresponders was 40 %, whereas that of CT volumetry was 73.3% and 3 Dimensional assessment was 60%, whereas specificity was 100 % for all three when clinicopathological downstaging was used as the reference standard. The agreement (weighted kappa) between RECIST and clinicopathological downstaging to classify the group of patients into complete response, partial response, stable and progressive disease were 0.66, whereas that of CT volumetry was 0.83 and that of 3 Dimensional assessment was 0.73.Using Bland Altman plot, the systematic bias of using RECIST criteria when compared to CT volumetry was 5.5%. The variance of the difference between the two methods was 9.85. The absolute percentage error was 39.04%. The systematic bias of using 3 Dimensional assessment when compared to CT Volumetry was 7.38%. The variance of the difference between the two methods was 7.19. The absolute percentage error was 35.38%.Conclusion Sensitivity of RECIST to detect responders was significantly low as compared to CT Volumetry. There was more agreement between CT Volumetry and clinicopathological assessment compared to RECIST, in categorisation of response assessment. Hence, CT volumetry may definitely be a better method than RECIST in assessing response to NACT in oral cavity tumors. Bland Altman plot analysis showed that there was slightly better agreement between 3 Dimensional assessment and volumetry as compared to RECIST. 11. Bland Altman plot analysis showed that there was slightly better agreement between 3 Dimensional assessment and volumetry as compared to RECIST. However the difference was not significant and hence 3 Dimensional measurement cannot be used as a substitute for volumetry in response assessment and further studies with more sample size needs to be done to assess this.

Keywords : Carcinoma Oral Cavity, RECIST, CT Volumetry, Three Dimensional Assessment, Clinicopathological Staging.