Previously, complex treatments plans could take more than a day to develop manually. Now, not only do the plans take a fraction of the time to create, but a much improved plan is produced that better protects sensitive tissues and other organs.
In addition, this new way of working has led to some patients having their dose of radiation to healthy tissue reduced by 20 to 30 per cent. This substantially lessens the probability of adverse reactions to the patient while having no detrimental effect on the effectiveness of the treatment.
The Beatson’s Radiotherapy Physics team was the first in the world to apply new and revolutionary multi-criteria optimisation (MCO) software which allows the planner to produce multiple treatment plans and determine the optimal and individualised solution for that patient. At least 20 to 30, and sometimes over 50, potential treatment plans for the patient are produced. This new system works in combination with a second suite of advanced software, based on a form of artificial intelligence. The software called RapidPlan, applies a mathematical model that is based on detailed planning information from previous similar treatments.
The models are either built by the Beatson team based on their local knowledge and experience, or based on models shared by other worldwide centres. Models at the Beatson have been produced for a wide range of treatments including head and neck, gynaecological and prostate cancer diagnoses.
The mathematical algorithm used in these software packages assimilates a vast amount of complex information and assists the planner in selecting the best plan. The human brain isn’t capable of processing all of this information in real time and counterbalancing the difficulties in obtaining the highest dose to the tumour while safeguarding doses to healthy tissues. For treatments that are highly complex, there can be almost an infinite number of solutions.
In addition to the speed in which treatment plans are developed, there are other unexpected benefits for patients and clinicians. This new way of working has, in some instances, able to reduce a patient’s chance of having cardiac issues in later life by 25 per cent compared to previous methods, by reducing dose to the heart.
Garry Currie, head of radiotherapy physics, said: “I didn’t expect us to reach this stage during my career. Our work never stands still; however this is a real game changer. As recently as 2010, staff would start work on just one treatment plan and come back in the next day – including weekends – to finish the calculations.
“Now we can input our own learning and combine it with others across the world to really tailor the best treatment plan for each patient. Around 200 of our patients are now benefitting from these rapid and high quality treatment plans every month. We can also produce multiple plans in advance of the patient coming into the department and adapt their treatment to best match their internal anatomy each day of treatment.
“We’re one of the biggest radiotherapy physics team in the world; however we’re also a very proactive team despite our size. We very quickly saw the benefits of MCO and the difference it could have to our patients and secured the licence as a result of obtaining Scottish Government capital funding for radiotherapy equipment. We’re also sharing our learning with colleagues around the world. This helps them advance their work while further building the global database of treatment plans that the AI software learns from.
“We have learned a number of things that we would never have expected. Now that we can plan in this way, we can effectively sculpt the dose and reduce the risk of damage to other organs. We now know that for some of our most advanced and complex plans, where we are treating large areas of the body, balancing the dose to one part, for example in the head region, can have a significant reduction in dose delivered much further away. Having these multiple plans created for the planner provides this excellent insight – in this case, this was something we didn’t previously realise or expect.
“We can now analyse these trade-offs in a dynamic fashion. Where we are treating a patient with a tumour in the pancreas, we can instantly assess the impact the plan would have on the dose delivered to either kidney and select the optimal balance in dose for that patient. We’re at the forefront of this work; however the results will continue to improve as more cancer centres start working in the same way and more and more collaborations take place.
“Even two years ago I didn’t forsee us working this way; we’re now able to share and collaborate, learning with UK teams and centres as widespread as Australia and Germany. As a result, the way we treat patients with radiotherapy can only improve and advance as this new way of working and the sharing of knowledge develops.”
Links