Calendar

<< Sept. 2010
Mo Tu We Th Fr Sa Su
    01 02 03 04 05
06 07 08 09 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      

The mother project (MADO)


 

Results of a European project (MADO): Organisational and economic implications of HLA diversity for HSC Donor Registries

« Optimisation of typing policies for European MArrow DOnor Registries: socio-economic evaluation of molecular techniques & recruitment strategies »
Research project, 5 EU countries, 14 partners, 3.5 years (2002-2005)
FP5 - QLG7-CT-2001-00065
Co-ordinator: Anne Cambon-Thomsen, Toulouse

 

Hematopoietic stem cell (HSC) donor Registries optimization in terms of HLA distribution and relevant recruitment strategies is a key objective for better service to patients.  It requires taking into account population HLA distribution and economical constraints, as HLA is very polymorphic and HLA typing has a relatively high cost. Interconnected National Registries should tend to be organized in an optimal way in order to increase the probability for a patient to find a donor. A theoretical work can be useful in this context to support action. The model presented was developed as part of the European Commission funded project MADO. A Registry model is characterised by the HLA phenotypes distribution in the population, the size of the registry and the sampling distribution of the donors. The efficiency of a registry is measured by the frequency of patients (recipients) who find a compatible donor. Given an initial registry and the size of its increment, we derived the optimal sampling scheme of the new donors and the maximal efficiency of the new registry. In this framework we obtain the following results:

 

  • 1. The maximum efficiency of a registry depends on the number of types in the population and of the geometric mean of their frequencies.
  • 2. An optimal sampling of donors should under recruit the ‘‘very frequent’’ phenotypes and the rarest ones while intermediate frequency phenotypes should be over represented compared to the population distribution.
  • 3. The difference between efficiency with random arrival and optimal selection of donors is quantified.
  • 4. The relation between size of registry and efficiency is theoretically derived and optimal size of this registry may be computed.
  • 5. The role of the probability for a compatible donor to be available for a transplant is taken into account.
  • 6. Efficiency of realisable optimal registry (based on pre-test of the donor) is considered.

 

As typing has a relative high cost, registry should be organized in an optimal way: What would be the optimal registry and how filtering strategy should be designed in order to improve current registry content?
In order to get decision elements, we define a welfare function associated with the existence of Registries, as the sum of the gain for patients and that for histocompatibility laboratories. We suggest a way to formalize efficiency of Registries and we apply it to the French HSC donor Registry. The main parameters of the model are the number of different HLA phenotypes in the population, their frequency distribution and the probability for a donor compatible at generic level with a patient, to be compatible at specific level and actually available for the graft, denoted a. The model determines an implicit value for the graft V = 55 831 euro and shows that the efficiency increases very slowly with the increasing number of potential donors in Registries. However the present national French recruitment program with the objective of recruiting 100 000 national new volunteer donors over 10 years appears reasonable.
The increase in size of Registries, in terms of number of donors, has only a marginal impact on efficiency, in an increase range foreseeable in France. The value of parameter “a” has a dramatic influence on the efficiency. Any decrease in the marginal cost would improve the welfare function. Such results can be of help in planning the evolution of Registries.

In order to test whether some kind of selection among donors presenting for becoming potential donors could be applied using genetic markers of the HLA region, we tested the following.
Because of linkage disequilibrium between microsatellites and HLA genes, we studied their possible use in Registry strategy. We wanted to evaluate whether MSat, that cost less than HLA, could help predict among candidate donors, those with HLA types most likely to optimise the use of Registries. Test hypothesis:  predicting the most frequent HLA types should be possible through MSat typing and by avoiding them, the genetic diversity of Registries would increase, consequently improving the probability of transplanting patients. To test this “molecular filter” we first chose 16 HLA-A, B, DRB1 allele level haplotypes, as being the most frequent ones in Europe. We then gathered ~7000 HLA typed DNA from potential donors mainly from France (~5000), but also from Italy, Hungary and Netherlands and had them typed for 16 MSat spanning the HLA region. In parallel theoretical models were constructed to reveal the parameters that allow an optimisation of Registries. The results show that:

  1. 1. multiple combinations of 3 or 4 MSat can be used to predict common HLA haplotypes;
  2. 2. it is better to estimate first MSat haplotypes to then predict combinations of HLA haplotypes;
  3. 3. the predictive power of various MSat combinations towards MSat haplotypes is not parallel to their ability to predict HLA haplotypes;
  4. 4. increasing immunogenetic diversity through avoiding very common HLA types does not optimally increase the potential use of Registries;
  5. 5. both the very frequent and the rare types should be avoided, as rare types also have a very low probability to get used. Present analyses are exploring the properties of MSat as indicators of degree of rarity of HLA types (without precisely predicting them). This degree of rarity is proposed as a decisional criterion for continuing or not HLA typing. This is subject to evaluation and proposed for discussion.

 

 In conclusion the interdisciplinary project MADO re-inforces the ideas that :

 

  1. 1. multiple approaches must be combined to increase the number of HSC transplantation;
  2. 2. some patients should be proposed other therapy rather than expecting finding compatible donors for very rare types;
  3. 3. the efficiency of the Registries is not synonymous of increased HLA diversity through rare types;
  4. 4. mechanism of optimisation of donor recruitment must be found;
  5. 5. the “filtration” of common HLA types using HLA microsatellites is not sufficient to increase notably the Registry efficiency;
  6. 6. decreasing the cost of typing and recruitment may have a dramatic effect. Applying a theoretical model of registry to plan future recruitment may be a good tool for helping decision in the general context of Public Health management.

 

MADO Workpackages