Without structured real world data, it remains guesswork who truly benefits  

The decision by the Dutch National Health Care Institute (Zorginstituut Nederland, ZiNL) to reassess the reimbursement of expensive medicines after their introduction into practice ,is a good first step toward structured evaluation. At the same time, it exposes a more fundamental problem: in practice, we often do not know for which patients expensive cancer medicines do or do not work. According to Hartwig Medical Foundation, this is because essential real-world data – such as genetic characteristics and treatment outcomes – are still not being recorded and used sufficiently. 

“If we want patients to receive the treatment that works for them, we must learn from real-world practice,” says Hans van Snellenberg, managing director of Hartwig. ‘ But at present, the necessary data infrastructure is missing. This limits our ability to provide appropriate care and leads to unnecessary costs.’  

Crucial information on treatment outcomes is often unavailable. In addition, much of the data is scattered across different sources that are barely connected. ‘ In the Netherlands, we have valuable information – including clinical data and genetic profiles – from sources such as the Netherlands Cancer Registry (NKR), PALGA, DICA, VEKTIS and Hartwig itself,” Van Snellenberg explains. “But as long as this information is not systematically linked and used as a coherent whole, there will remain uncertainty about what truly works for which patient. ‘  

Personalised treatment is not a luxury, but a necessity 
The need is growing. According to the latest Horizonscan Medicines, 137 new cancer drugs are expected in the next two years, many of them targeting specific molecular features. This makes a personalised approach essential – but without real-world insights, it remains difficult to determine which patient will benefit from which drug. 

Patient organisations, such as the Dutch Federation of Cancer Patients’ Organisations (NFK), also stress that real-world data is crucial to understanding what does work. In the current reassessment of PARP inhibitors, such data is lacking, NFK notlNK/ed in its response. “It’s a first step that the Health Care Institute, in its advice, alsom calls for ensuring that real-world data becomes available and that initial assessments should already indicate which data needs to be collected. As patient organisations, we will strongly advocate to ensure that this actually happens in the future.” 

Time for systemic change 
ZiNL’s reassessment highlights the urgency of structural change. “Medicines must be monitored more effectively from the moment they are approved, based on real-world data,” says Van Snellenberg. “Only then can future reassessments be carried out with a stronger evidence base. This is, above all, in the patient’s interest, but it also prevents the waste of healthcare funds.” 

In 2024, the Netherlands spent around €2.7 billion on oncology medicines. Estimates suggest that a significant share of this went to treatments that are only marginally effective. “Even a modest improvement in cost-effectiveness could save hundreds of millions – money we could then reinvest in medicines that do work,” Van Snellenberg concludes. 

You read an article in the category Personalized treatment. You may also be interested in Billing code, Hartwig Medical Database, Learning healthcare system All news