Oncology

A recent high-impact study was the update of DRG prices for autologous and allogeneic stem cell transplantation. For autologous stem cell transplantation, inpatient days turned out to be the key cost driver. For allogeneic stem cell transplantation from unrelated donor, costs of selecting and harvesting the stem cells were cost-drivers together with inpatient days.

We are proud to share the following research: iMTA developed full disease models for economic evaluations in multiple myeloma and renal cancer that are able to identify effects of second-line treatments, while including effects of first-line treatments. The strategy we employed, was developing a patient-level data model, rather than modelling a cohort using a Markov Model. The model has been made applicable to other types of cancer.

In the new area of personalized medicine, identifying patients in which medication is optimally effective is essential. iMTA worked on early HTA for diagnostic tests for acute leukemia and multiple myeloma, two life-threatening types of blood cancer, lung cancer, colon cancer and head & neck cancer. The challenge is to identify the different subgroups of disease, based on genetic differences in the tumors to help optimize treatment. iMTAs early HTA helped both in developing and marketing the tested biochip by identifying cost-savings and modelling health gains that followed better targeted treatment.

We have initiated the set up of several population based, observational registries, a.o. PHAROS in hematology, PERCEPTION in renal cancer and CAPRI in castration-resistent prostate cancer. The aim of these registries are to get insight in the impact of new treatments of patients with malignancies in the Netherlands in terms of process and outcomes (including cost-effectiveness). Real world data has many particular challenges which we tried to solve using decision analytic modelling. Our approach besides developing well-targeted models: invest time in study design, use propensity score matching methods to identify a good comparator and try different multiple imputation methods for missing data.

iMTA currently participates in the organization of six oncology registries for prostate -, lung -, renal - and colon cancer, melanoma and hematology. iMTA guides and performs the cost-effectiveness track of these registries, as well as an assessment of real life drug uptake and prescription behavior. Analyses show that there is regional variation in prescription behavior, resulting in some patients not receiving indicated treatment with innovative medicines.

What we often see when performing a systematic review in cost-effectiveness studies in cancer: studies are often not generalizable and use a too short horizon. For this reason, we internally work with a quality assurance system for our health economic models.

iMTA faced and solved many issues in analytic decision-making in cancer studies. A particular challenge was dealing with real world data: patients with unique treatment paths, large heterogeneity in both patient characteristics and treatment and missing data. Our approach: identify uncertainty, patient-level models rather than Markov models, smart simulations to deal with patient heterogeneity and correct for treatment heterogeneity where possible. Currently successfully employed in real-world cost-effectiveness models in follicular lymphoma and renal cancer.

The first landmark reimbursement dossier for late phase study with real world data was for Oxilaplatin. The cost-effectiveness study was difficult, due to large patient and treatment heterogeneity. Modelling efforts were successful and indicated an incremental cost-effectivenss ratio of about 10,000 euro per QALY for oxilaplation plus fluoropyrimidines (FL) compared to only FL. We later applied the lessons to numerous other reimbursement dossiers for late phase studies, including for multiple myeloma and non-Hodgkin’s lymphoma. In all outcomes and reimbursement studies, iMTA also looks at real world drug use. More often than not, we see slow uptake of innovative medicines.

iMTA developed the condition specific QLQ-PBM, a preference based instrument based on the QLQ-C30. Using the Time Trade-off method in about 400 group interviews in The Netherlands, a selection of items (selected using Rasch-analysis and expert opinion) of the QLQ-C30 was weighted. The resulting instrument is more sensitive than EQ5D for minor impairments. Utilities can be calculated from the QLQ-C30 directly. iMTA also developed several 'mapping algorithms' to estimate EQ5D utilities from QLQ-C30.

Reimbursement dossiers
The first landmark reimbursement dossier for late phase study with real world data was for Oxilaplatin. The cost-effectiveness study was difficult, due to large patient and treatment heterogeneity. Modelling efforts were successful and indicated an incremental cost-effectivenss ratio of about 10,000 euro per QALY for oxilaplation plus fluoropyrimidines (FL) compared to only FL. We later applied the lessons to numerous other reimbursement dossiers for late phase studies, including for multiple myeloma and Non-Hodgkin. In all outcomes and reimbursement studies, iMTA also looks at real world drug use. More often than not, we see slow uptake of innovative medicines.