1

Introduction

The impact of Breast Conserving therapy [BCT] has been monumental on quality of life. Assessment of Cosmesis and Quality of life forms an important endpoint for clinical research. Various studies have demonstrated high correlation between poor cosmetic outcome and low scores on quality of life indicators [ , ].

Subjective or objective cosmetic evaluation is integral to clinical trials related to breast conserving therapy. Subjective assessment may be done utilizing the Harris scale [ ] The Harris scale reports the outcome on a 4 point-Likert scale [excellent, good, fair and poor] based on the degree of asymmetry between the treated and untreated breasts. Even though the Harris scale is the simplest method used to subjectively score cosmesis, it is prone to high variability due to various factors like the professional background of the rater, method of capturing and processing the image, etc.

The BCCT.core is a commonly employed software for objective evaluation that processes images and rates the cosmesis on a 4-point scale [Excellent, Good, Fair or Poor] using a semi-automatic software utilizing predetermined points placed on a frontal image. It is developed and owned by Cardoso J, Cardoso M and INESC Porto Breast Research Group. The main objective was to develop a reproducible and widely available methodology for evaluation of aesthetic results in breast cancer conservative treatment, enabling effective comparison of outcome between centers [ ].

The BCCT.core software is easy to use and user friendly. It requires an upright breast photograph with a light background and good contrast resolution. The outcome of the assessment is based on various different factors, predominantly the asymmetry between the treated and untreated breasts. The software utilizes fixed points on the image (sternal notch, and a fixed distance below the sternal notch), and breast contour as set by the user on the image, to give a global cosmetic rating. An illustration of the software is shown in the supplementary file as figure no. 1.

Breast Reconstruction Assessment [BRA] is also an objective method of cosmetic evaluation [ ]. However, all of these assessment scales come with inherent inconsistency and bias. Pezner et al. have demonstrated that it is difficult to obtain consensus of cosmetic outcome with the commonly employed scales [ ].

Cosmetic rating on breast photographs is commonly done in clinical trials, mostly retrospectively by a panel of experts like in the UK START Trial B [ ]. If the rating is done using breast photographs, factors affecting the image quality will also affect the cosmetic rating. Use of standardized methods to acquire the photographs under uniform room lighting conditions, distance of the camera from the patient, and specifications of the camera can minimize this variability. The rating done by a clinician is also often different from the rating done by patients [ ]. It also varies with the experience [ ], and fatigue of the rater [when a bulk of images have to be rated].

It is also possible that if a clinician rates the images in serial order chronologically, one patient at a time, the rating of consequent images might change after rating the first image, either negatively or positively, thereby creating a bias. It is noteworthy that routine cosmetic evaluation done in high volume clinics is random since the clinician is unlikely to compare the current cosmetic status with a previously taken image. Hence, when cosmetic analysis is done retrospectively, and images of the patient taken at different time points are available to influence the decision of the rater, the resultant rating might be different from actual assessments done in clinics [which are random]. At the same time, serial images may also guide the clinician to give a better informed global cosmetic rating to ultimately know the temporal changes [worsening or improvement] in the treated breast.

In an IEC approved study being conducted in our center [CTRI/2020/01/022871], images of breast cancer patients who underwent breast conserving surgery [BCS] and adjuvant Radiotherapy have been collected at baseline and follow up visits, for cosmetic evaluation by a panel. These images were utilized in the current study with the objective of evaluating inter and intra-rater bias for assessment of cosmesis in the serial vs. random assessment settings, and if the experience of the rater impacts the global cosmetic rating.

2

Materials and Method