Breast Cancer
Conditions
Brief summary
The aim of this study is to develop a computer model, based on photographs and heat images of patients' skin, to provide early prediction of painful reddening of the skin during radiotherapy treatment.
Detailed description
There is currently no reliable tool to quantify and detect erythema of the skin during radiotherapy. This side-effect may lead to painful moist desquamation, and eventually permanent delayed side effects like telangiectasia. If such a tool would be available, several interventions could be staged, including (1) the use of steroid cream\], (2) the re-simulation and/or re-planning of patients to decrease the skin dose by spreading out the entrance of the beams, (3) adjusting or eliminating the use of bolus on the skin surface (which boosts superficial dose) or (4) the use of other treatment techniques including prone technique. Given that the dose delivered is not a reliable metric to predict for erythema in a given patient, a new method for monitoring, staging and ultimately predicting skin response is needed. By analyzing images of the skin using both visible and infrared spectral regions, and by carefully converting the information in the images to quantitative metrics, it may be possible to characterize the stage of a patient's response to radiation, and to understand which patients may go on to experience chronic pain, severe burns or other more serious side effects while it is still early enough to intervene. The proposed research is to develop a software model that will take as input patient skin image data and the patient known clinical outcomes and algorithmically generalize a model to predict a biological response of skin to ionizing radiation for any future patient, after a few initial images. In the first stage of this study, the data will be aggregated to devise the dose response curve. In later phases, the model will be refined and used for predictive purposes, i.e., once a new patient has begun radiotherapy sessions, their initial response will be quantified, and fed into the model to predict the skin response endpoint after the course of radiation therapy ends. As mentioned, this information could be used to adapt the radiation course and optimize the therapy for the individual, potentially preventing morbidity from overdose, or risk of recurrence from under dose.
Interventions
OTHEROptical and infrared imaging
Acquiring images (photographs) of the skin
Sponsors
Nova Scotia Health Authority
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
PREVENTION
Masking
NONE
Eligibility
Sex/Gender
FEMALE
Healthy volunteers
No
Inclusion criteria
\- All female patients referred for adjuvant radiotherapy following breast conserving surgery/mastectomy.
Exclusion criteria
* Patients with known skin issues (e.g. dermatomyositis, rosacea) * Patients with excessive risk of skin recurrence, including T4d * Patients with locally advanced breast cancer.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Establish the correlation between optical and IR skin imaging and skin toxicity | Week one of radiation treatment | Acute skin side effects assessed using the National Cancer Institute of Canada Common Terminology Criteria for Adverse Events, version 4.03. Scale ranges from 1 to 5. Correlation will be performed relative to optical image colour, texture, and temperature reflected by IR imaging. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Establish the correlation between optical and IR skin imaging and skin itchiness and pain. | Week one of radiation treatment | Weekly quantitative visual analog scale (VAS) measures of itchiness and/or pain Scale ranges from 0 (no itchiness or pain) to 10 (extreme itchiness or pain). Correlation will be performed relative to optical image colour, texture, and temperature reflected by IR imaging. |
| Establish the correlation between optical and IR skin and Quality Of Life (QOL). | Week one of radiation treatment. | The European Organization for Research in Treatment of Cancer (EORTC) Quality of Life Questionnaire (QLQ)-C30 and -BR23 will be used (30 questions, scale of 1 to 4). Correlation will be performed relative to optical image colour, texture, and temperature reflected by IR imaging. |
Countries
Canada
Outcome results
None listed