ARCHERY

A prospective observational study of artificial intelligence based radiotherapy treatment planning for cervical, head and neck and prostate cancer

Can artificial intelligence (AI) efficiently plan radiotherapy treatment for head and neck, prostate, and cervical cancers?

What is this study about?

Half of patients with cancer require radiotherapy, however, in low and middle-income countries (LMICs) only 10-40% of patients have access to this treatment. Shortages of equipment and in specialised workforce are the most significant barriers to expanding the availability of radiotherapy in LMICs. In addition, the current radiotherapy planning pathway, which is a central part of the radiotherapy care pathway, is inefficient. The process of taking images, deciding where to direct the radiotherapy and creating a treatment plan is time-consuming.

Currently, these labour-intensive steps can take weeks to months to deliver in LMICs. This contributes to long waiting times for treatment and can result in progression (where the cancer continues to grow), increased morbidity (complications resulting from the cancer growing), and worse survival. In addition, inaccuracies at each stage of radiotherapy planning pathway can directly impact patients in terms of survival and side effects from treatment.

The ARCHERY study aims to test a new AI-based software that can automate two components of the radiotherapy planning pathway:

Identify and draw on a CT scan the areas that are at risk of cancer spread and protect those body parts at risk of radiation damage.
Decide on the position, size, and shape of the radiation beams to treat the cancer effectively and minimise radiation to areas at risk of radiation damage.

This automation has the potential to improve radiotherapy in three ways:

reducing the time needed to outline treatment structures and create radiation plans from hours to minutes
reducing variation between clinicians and improving the quality of treatment plans
supporting better use of expert human resources

In this study the AI computer software uses participants' CT scans only. There are no clinical impacts from the intervention or risk to the participants as they will continue to be treated according to the standard treatment pathway. There is no additional imaging or treatment. Furthermore, participants will not require any additional study visits.

ARCHERY is a 4-year prospective observational study to evaluate the quality and economic benefit of AI-based automated radiotherapy treatment planning. The study focuses on cervical, head and neck, and prostate cancers, which are common in LMICs, and for which radiotherapy is the main curative treatment. The study will recruit 990 patients across six public-sector cancer hospitals in India, Jordan, Malaysia, and South Africa. Including a broad range of patients from these countries will ensure the findings are representative, supporting the implementation of the software across all LMICs.

Partners from the patient community have informed the scope, objectives, and design of the study. Patients will also advise throughout the study life cycle, including study conduct, and co-design of the dissemination of results and public engagement activities.

If the study objectives are met, the AI-based software will be offered as a not-for-profit web service to public sector state hospitals in LMICs. We anticipate this will support rapid adoption into daily practice and facilitate the expansion of high-quality radiotherapy in these countries. This, in turn, will improve access and affordability of this critical treatment option, and therefore also improve survival and the quality of life of patients with cancer globally.

ARCHERY is led by Dr Ajay Aggarwal, a Consultant Clinical Oncologist at Guy's & St Thomas' NHS Trust in London and an Associate Professor at the London School of Hygiene and Tropical Medicine and King's College London.