Skip to content

Testing a Web-Based Chatbot to Help Newly Diagnosed Breast Cancer Patients Understand Their Treatment Options

Pilot Trial of a Hybrid Decision Aid Chatbot Combining Decision Logic and Large Language Model for Personalised Breast Cancer Treatment Planning in Newly Diagnosed Breast Cancer Patients

Status
Not yet recruiting
Phases
Unknown
Study type
Interventional
Source
ANZCTR
Registry ID
ACTRN12625000448448
Enrollment
60
Registered
2025-05-13
Start date
2027-06-28
Completion date
2027-12-31
Last updated
2026-07-13

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

None listed

Brief summary

This study evaluates a hybrid decision-support chatbot designed to assist newly diagnosed breast cancer patients in understanding and planning their treatment. Who is it for? You may be eligible to participate in this study if you are a female aged 18 or over, invasive breast cancer (any histologic subtype) and have access to a device with internet access. Study details Participants in this study are randomly allocated by chance (similar to flipping a coin) to one of two groups. Participants in one group will use a chatbot at the time of diagnosis or treatment planning to enter their clinical data and receive a personalised treatment summary, along with standard care. The chatbot integrates a rule-based decision engine based on international clinical guidelines (e.g., NCCN, ESMO, eviQ, ASCO) with a large language model (LLM) interface to generate personalised treatment summaries in plain language. The chatbot provides information on surgical options, systemic therapy, treatment sequencing, and adjuvant therapies. It is intended to support shared decision-making and improve patient understanding but does not replace clinical judgment. All treatment decisions remain under multidisciplinary team (MDT) supervision. Participants in the other group will receive standard care alone which includes review of diagnostic imaging and pathology, MDT discussion and treatment recommendation, consultation with a breast surgeon and/or oncologist, and provision of verbal and written information about treatment options. Primary outcomes include feasibility (recruitment and retention rates) and concordance between chatbot and MDT recommendations. Secondary outcomes include usability (System Usability Scale), decision confidence (Decision Conflict Scale), and patient acceptability, satisfaction, and engagement (via thematic analysis of semi-structured interviews).

Interventions

This is a prospective, single-arm, interventional pilot study evaluating the use of a mobile phone application designed to support treatment planning for newly diagnosed breast cancer patients. The intervention involves providing participants with access to a decision-support app that generates personalized treatment recommendations based on user-entered clinical and pathological data, including tumor size, location, receptor status (Estrogen Receptor [ER], Progesterone Receptor [PR], Human Epid

This is a prospective, single-arm, interventional pilot study evaluating the use of a mobile phone application designed to support treatment planning for newly diagnosed breast cancer patients. The intervention involves providing participants with access to a decision-support app that generates personalized treatment recommendations based on user-entered clinical and pathological data, including tumor size, location, receptor status (Estrogen Receptor [ER], Progesterone Receptor [PR], Human Epidermal Growth Factor Receptor 2 [HER2]), nodal involvement, and patient demographics. Participants will use the app at the time of diagnosis or treatment planning, entering their own clinical information (with assistance from a study coordinator if needed). The app will generate a personalized summary, which includes: • Suitability for breast-conserving surgery • Likelihood of requiring sentinel lymph node biopsy or axillary dissection • Likelihood of needing neoadjuvant chemotherapy ± trastuzumab • Predicted sequence of treatment (e.g., neoadjuvant vs. upfront surgery) • Likelihood of requiring adjuvant therapies (chemotherapy, radiotherapy, endocrine therapy) This summary is provided solely for informational purposes to assist patients in understanding potential treatment options. The app does not provide medical advice or replace clinical judgment, and treatment decisions remain under the direction of the multidisciplinary team (MDT). The app’s outputs will be compared to actual MDT decisions to assess concordance. As participants only receive access to the app and its outputs by enrolling in this study, this will be considered a non-interventional study. The application summary is intended for informational purposes only and is not used for direct decision-making alongside the MDT. Study Duration and Frequency: The app will be used once at the time of diagnosis or treatment planning, with an expected duration of approximately 30-60 minutes for data entry and review. Strategies to Assess Adherence: Adherence to the intervention will be assessed using app analytics, including usage rates, time spent on the app, and completion of modules.

This is a prospective, single-arm, interventional pilot study evaluating the use of a web-based chatbot designed to support treatment planning for newly diagnosed breast cancer patients. The intervention involves providing participants with access to a decision-support chatbot that generates personalised treatment recommendations based on user-entered clinical and pathological data, including tumor size, location, receptor status (Estrogen Receptor [ER], Progesterone Receptor [PR], Human Epiderm

This is a prospective, single-arm, interventional pilot study evaluating the use of a web-based chatbot designed to support treatment planning for newly diagnosed breast cancer patients. The intervention involves providing participants with access to a decision-support chatbot that generates personalised treatment recommendations based on user-entered clinical and pathological data, including tumor size, location, receptor status (Estrogen Receptor [ER], Progesterone Receptor [PR], Human Epidermal Growth Factor Receptor 2 [HER2]), nodal involvement, and patient demographics. The chatbot operates in two stages: 1. Initial Diagnostic Stage: At the time of diagnosis or initial treatment planning, participants enter their clinical information into the chatbot. Based on this input, the chatbot provides: - A preliminary diagnosis summary - Suitability for breast-conserving surgery - Likelihood of requiring sentinel lymph node biopsy or axillary dissection - Likelihood of needing neoadjuvant chemotherapy ± other systemic therapies - Predicted treatment sequence (e.g. neoadjuvant vs. upfront surgery) - Likelihood of requiring adjuvant therapies (chemotherapy, radiotherapy, endocrine therapy) 2. Postoperative Stage: After surgery, participants (or study coordinators) enter updated histopathology results. The chatbot then generates a definitive treatment recommendation, incorporating postoperative findings to align with multidisciplinary team (MDT) decision-making. This summary is provided solely for informational purposes to assist patients in understanding potential treatment options. The chatbot does not provide medical advice or replace clinical judgment, and treatment decisions remain under the direction of the MDT. The chatbot’s outputs will be compared to actual MDT decisions to assess concordance. As participants only receive access to the chatbot and its outputs by enrolling in this study, this will be considered a non-interventional study. The chatbot summary is intended for informational purposes only and is not used for direct decision-making alongside the MDT. Study Duration and Frequency: The chatbot will be used twice: - Once at the time of diagnosis or initial treatment planning - Once after surgery, incorporating histopathology results?Each session is expected to take approximately 30–60 minutes for data entry and review. Strategies to Assess Adherence: Adherence to the intervention will be assessed using chatbot analytics, including usage rates, time spent on the chatbot, and completion of modules.

Sponsors

Chris O'Brien Lifehouse
Lead SponsorHospital

Study design

Allocation
Randomised controlled trial
Intervention model
Parallel
Primary purpose
Educational / counselling / training
Masking
Open (masking not used)

Eligibility

Sex/Gender
All
Age
18 Years to No maximum
Healthy volunteers
Yes

Inclusion criteria

Participants must meet all of the following criteria: For patients: Aged 18 years or older Newly diagnosed with invasive breast cancer (any histologic subtype) Planned for treatment with curative intent within a multidisciplinary team setting Able to provide informed consent Have access to a smartphone or tablet device Able to read and understand English (or the language in which the app is available) For clinicians: Involved in the care of breast cancer patients within the participating multidisciplinary team Have used or reviewed the decision-support app as part of the study Willing and able to provide feedback through interviews or surveys Able to provide informed consent

Exclusion criteria

Participants will be excluded if they meet any of the following criteria: For patients: Diagnosis of non-invasive breast cancer only (e.g., ductal carcinoma in situ [DCIS] without an invasive component) Evidence of metastatic (stage IV) breast cancer at diagnosis Enrolled in another clinical trial that could influence treatment planning Cognitive impairment or other condition affecting the ability to use the app or provide informed consent No access to mobile technology or a reliable internet connection For clinicians: There are no specific exclusion criteria for clinicians participating in the feedback component of the study.

Outcome results

None listed

Source: ANZCTR · Data processed: Jul 23, 2026