Digital twins and prevention : a new era in medicine

Digi­tal twins are gai­ning ground in the indus­trial world. The doors that this kind of tool opens up allow it to be exten­ded into the field of heal­th­care ; in par­ti­cu­lar with the use of digi­tal lungs for per­so­na­li­sed care. Being able to model an organ, per­so­na­lise this model using patient data, and simu­late its func­tio­ning could lead us to plan, anti­ci­pate and pre­dict its deve­lop­ment and ageing.

After all, as Claire Biot, Vice-Pre­sident of the Health Indus­try at Das­sault Sys­tèmes, right­ly points out, “today, by having the digi­tal twin of a Rafale, as well as lots of flight data – the equi­va­lent of bio­mar­kers in aero­nau­tics – we can car­ry out pre­dic­tive main­te­nance in ope­ra­tion. This means moving from the digi­tal twin of a concept, the air­craft, to the per­so­na­li­sed digi­tal twin of each Rafale in operation.”

This raises the ques­tion : what role can digi­tal twins play in pre­ven­tive heal­th­care ? A ques­tion that was at the heart of the dis­cus­sions at the 5^th semi­nar of the Poly­tech­nique health group, coor­di­na­ted by the Centre de recherche en ges­tion de l’École poly­tech­nique (IP Paris), and led by Étienne Min­vielle (CNRS Research Direc­tor)¹.

Health pre­ven­tion is often seen as a sub­ject that was neglec­ted until recent­ly. Stan­ley Durr­le­man, research direc­tor at Inria, is quick to point this out. “Tal­king about pre­ven­tion is still uncom­mon in France, because first and fore­most we have a care sys­tem rather than a health sys­tem.” In his opi­nion, the cur­rent mot­to in research is more along the lines of “bet­ter to cure than to prevent.”

At Das­sault Sys­tèmes, digi­tal twins have been in use for some time. “Our ambi­tion is to help people live heal­thier lives at a cost that is sus­tai­nable for heal­th­care sys­tems, and we place digi­tal twins at the heart of our stra­te­gy,” explains Claire Biot, “because they make it pos­sible to represent the invi­sible and get seve­ral dis­ci­plines to work toge­ther.” A new pro­ject has been laun­ched : “MediT­win was laun­ched by Das­sault Sys­tèmes to create dif­ferent digi­tal twins in three the­ra­peu­tic areas : onco­lo­gy, neu­ro­lo­gy and car­dio­lo­gy,” she explains. “These digi­tal twins will pri­ma­ri­ly be used for treat­ment or diag­no­sis, gui­ding medi­cal decisions”.

There are two types of pre­ven­tion : pri­ma­ry pre­ven­tion (put in place to avoid cat­ching the disease) and secon­da­ry pre­ven­tion, which stops or slows down the deve­lop­ment of the disease. Pri­ma­ry pre­ven­tion applies in par­ti­cu­lar to com­mu­ni­cable diseases, with condoms and qua­ran­tine as examples. It is the­re­fore cer­tain that, for cer­tain diseases, pri­ma­ry pre­ven­tion is com­pli­ca­ted. The second form of pre­ven­tion has a num­ber of constraints, in par­ti­cu­lar the need for rapid diag­no­sis and long-term patient mana­ge­ment. “How can I fol­low patients for decades to see if, in the end, my pre­ven­tion mea­sure was effec­tive?” won­ders Stan­ley Durr­le­man. This is where a digi­tal model­ling sys­tem that fai­th­ful­ly simu­lates the func­tio­ning of a per­so­na­li­sed bio­lo­gi­cal sys­tem using patient data comes into its own.

When it comes to illus­tra­ting the com­ple­men­ta­ri­ty bet­ween pre­ven­tion and the pre­dic­tive algo­rith­mic sys­tems pro­po­sed by digi­tal twins, Stan­ley Durr­le­man consi­ders Alzheimer’s disease to be the prime example. Alzheimer’s disease is cur­rent­ly not far behind can­cer as one of the big­gest areas of medi­cal research. “Des­pite colos­sal invest­ments in research, we haven’t had a posi­tive cli­ni­cal trial for this disease for 20 years,” explains the resear­cher. “And prac­ti­tio­ners still have vir­tual­ly nothing to pres­cribe to their patients” [editor’s note : the drug in ques­tion is cur­rent­ly only avai­lable in the Uni­ted States, and the appli­ca­tion to mar­ket it in Europe is still being exa­mi­ned by the authorities].

This treat­ment tar­gets the accu­mu­la­tion of amy­loid plaques in the patient’s brain, which begins 10–15 years before the first symp­toms of the disease appear. “Howe­ver, the effects of the treat­ment on the patient’s cog­ni­tive decline remain rela­ti­ve­ly minor,” he admits. “The change in our treat­ment plan is sub­stan­tial. We are no lon­ger loo­king for drugs to cure the disease, but rather to prevent it.”

It is the­re­fore impor­tant to unders­tand how this disease deve­lops and what the first signs are, in order to make an ear­ly diag­no­sis. To this end, Stan­ley Durr­le­man’s research team has esta­bli­shed pre­dic­tive models – digi­tal twins – that can deter­mine the course of the disease in dif­ferent people. After pre­sen­ting this model, Claire Biot asked : “Are we rea­dy for mass-tes­ting for Alz­hei­mer’s disease?” Stan­ley Durr­le­man replies that there is still pro­gress to be made. “For a non-com­mu­ni­cable disease, it’s more accu­rate to talk about scree­ning,” explains the resear­cher. In gene­ral, for this type of ill­ness, the patient will a doc­tor fol­lo­wing the first symptoms.

Iden­ti­fi­ca­tion the­re­fore depends on the patient taking action, and this first step must not be mis­sed. “Given the issues we have alrea­dy dis­cus­sed, I think there is still a lot of work to be done to improve the detec­tion of cog­ni­tive disor­ders in the gene­ral popu­la­tion,” he concludes. One of the areas to be deve­lo­ped would be the iden­ti­fi­ca­tion of bio­mar­kers spe­ci­fic to the deve­lop­ment of Alz­hei­mer’s disease. Howe­ver, just as the accu­mu­la­tion of amy­loid plaques in a patient’s brain is not always syno­ny­mous with the first signs of the disease, pre­ven­tion will always depend on our abi­li­ty to predict.

These fin­dings confirm that, even if ques­tions remain, the digi­tal twin is like­ly to play an impor­tant role in the coming years. It could herald a new form of health pre­ven­tion, more algo­rith­mic, more per­so­na­li­sed and offe­ring pre­vious­ly uni­ma­gi­nable detec­tion capabilities.

Pablo Andres