Will personalised medicine create problems for the economy ?

You seem enthu­sias­tic about stu­dying how this new medi­cal approach is being imple­men­ted, and the unique ques­tions it’s rai­sing. You are also loo­king at the risks that it could pose for our cur­rent heal­th­care sys­tem. From an eco­no­mic stand­point, what are the stakes of per­so­na­li­sed medicine ?

It’s a new area of research for health eco­no­mists. It chal­lenges our tra­di­tio­nal fields of stu­dy – that is, the doc­tor-patient rela­tion­ship, access to heal­th­care and medi­cal and eco­no­mic eva­lua­tions of the­ra­peu­tic approaches [which deter­mine how social secu­ri­ty resources are allo­ca­ted, to pro­vide the best care to the population].

For us, per­so­na­li­sed medi­cine means three major changes. The first is the shift from a one-size-fits-all sys­tem to an indi­vi­dua­li­sed sys­tem, using a patient’s gene­tic infor­ma­tion. The second is the shift from a reac­tive approach, on the basis of symp­toms, to a proac­tive approach, which aims to anti­ci­pate and prevent diseases before they even occur. Final­ly, this kind of medi­cine uses mass data, pro­du­ced by new tools for DNA analysis.

Doc­tors are won­de­ring how to inter­pret these results and how they should be given to the patient, espe­cial­ly when they indi­cate a cer­tain pre­dis­po­si­tion (i.e. a stron­ger risk of deve­lo­ping a disease in the future). It’s also an indi­ca­tion on the way patients, civil socie­ty, and pro­fes­sio­nals pre­fer to com­mu­ni­cate gene­tic data.

How is this dif­ferent from tra­di­tio­nal medicine ?

In large part because of secon­da­ry data, which has nothing to do with the rea­son the patient came in for a consul­ta­tion in the first place. Gene­tic tes­ting often pro­duces secon­da­ry data indi­ca­ting a patient’s pre­dis­po­si­tions to other ill­nesses, with varying degrees of certainty.

In some cases, tes­ting shows a pre­dis­po­si­tion to a disease for which treat­ments or pre­ven­tion pro­to­cols can be imple­men­ted or the patient’s cli­ni­cal moni­to­ring can be adjus­ted. But some­times there is nothing to be done. A typi­cal example is Huntington’s disease (a rare, here­di­ta­ry neu­ro­de­ge­ne­ra­tive disease with no real treat­ment). These pre­dis­po­si­tions can also affect other mem­bers of the patient’s family.

Why are eco­no­mists interested ?

They want to know more about the unique doc­tor-patient rela­tion­ship, how we decide who can bene­fit from this gene­tic tes­ting as well as who can­not, and the­re­fore the ways in which this medi­cal approach can be acces­sed. In a pater­na­lis­tic sys­tem, the doc­tor makes the deci­sion. But in a sys­tem where deci­sion-making is sha­red more even­ly bet­ween the two par­ties, the doc­tor should allow the patient to choose their own pre­fe­rences about what they want to know and what they would rather not know. This can be impor­tant for eco­no­mic eva­lua­tion in geno­mic medicine.

But this secon­da­ry data is not neces­sa­ri­ly use­ful from a cli­ni­cal point of view ?

Repor­ting a patient’s secon­da­ry pre­dis­po­si­tions can lead to the imple­men­ta­tion of pre­ven­tion pro­to­cols, or modi­fi­ca­tion of the­ra­peu­tic moni­to­ring. For example, if we know that a patient is pre­dis­po­sed to a very aggres­sive kind of breast can­cer, we can recom­mend pre­ven­tive sur­ge­ry. The Ame­ri­can Col­lege of Medi­cal Gene­tics and Geno­mics recom­mends advi­sing patients of these kinds of pre­dis­po­si­tions, and keeps a regu­lar­ly upda­ted list of actio­nable genes, for which effec­tive treat­ment is possible.

In this context, it’s clear that it is in the public inter­est for the regu­la­tor to pro­vide access to this data. At the moment, it’s not autho­ri­sed. But dis­clo­sing this infor­ma­tion can affect an individual’s beha­viour. From a medi­co-eco­no­mic eva­lua­tion pers­pec­tive, this involves going beyond cli­ni­cal cri­te­ria. From the patient’s point of view, it can be valuable to access this secon­da­ry data, even when their genes are not actionable.

This is what’s cal­led data’s per­so­nal uti­li­ty – kno­wing can influence our choices. A diag­no­sis has a psy­cho­lo­gi­cal, plan­ning and cli­ni­cal value. Psy­cho­lo­gi­cal value means the intrin­sic value of the infor­ma­tion, the fact of kno­wing that you are pre­dis­po­sed to a cer­tain disease. Plan­ning refers to the way it can impact life choices, such as the deci­sion to have ano­ther child or to buy a property.

And in finan­cial terms ?

Beyond results, geno­mic medi­cine is also sha­king up the eco­no­mic side of things. For example, with rare diseases, gene­tic pro­fi­ling can replace a myriad of tests and avoid years of incor­rect diag­no­sis, the­re­fore resul­ting in savings.

It is very dif­fi­cult to assess the ave­rage cost of per­so­na­li­sed medi­cal tes­ting. Throu­ghout the world, attempts are being made to put a figure on this kind of care. In France, we know that the Natio­nal Health Autho­ri­ty (HAS) is trying to assess whe­ther per­so­na­li­sed medi­cine can be pro­vi­ded to patients for an accep­table cost. But the tech­no­lo­gy is expen­sive. Pur­suing it may eli­mi­nate the pos­si­bi­li­ty of adop­ting other stra­te­gies that are just as use­ful from a cli­ni­cal point of view, as per­so­na­li­sed medi­cine would take up a signi­fi­cant part of the limi­ted public bud­get. It’s some­thing one should keep in mind.

What shifts are you expec­ting to see from a heal­th­care path­way point of view ?

Nowa­days, geno­mic ana­ly­sis is the last step. It occurs only when the patient has been diag­no­sed and is seeing a spe­cia­list. But accor­ding to the France Geno­mic Medi­cine Plan, in the future, this ana­ly­sis could be used for com­mon diseases. Does this mean that one will need their own gene­tic data to receive heal­th­care ? That is not what gene­ti­cists are sug­ges­ting, but this is an impor­tant question.

Per­so­na­li­sed medi­cine has alrea­dy brought shifts. Mul­ti­dis­ci­pli­na­ry consul­ta­tions had to be crea­ted, as well as new pro­fes­sions, spe­ci­fi­cal­ly in bio­tech. This raises issues for edu­ca­tion and trai­ning policy.

We might also see shifts in the insu­rance sec­tor. Our insu­rance sys­tem is finan­ced col­lec­ti­ve­ly and based on the fact that indi­vi­dual risk is mas­ked by a veil of igno­rance. If, with this data, each person’s risk is revea­led, this could have conse­quences such as insu­rers refu­sing to cover some people. Insu­rers could also place indi­vi­dual res­pon­si­bi­li­ty on patients, man­da­ting beha­viou­ral change if cer­tain pre­dis­po­si­tions are iden­ti­fied. We alrea­dy know that pre­ven­tive beha­viours depend on a patient’s social envi­ron­ment. This means there is a risk that social health inequa­li­ties could widen if this approach becomes widespread.