Journal Logo

Commentaries

The 3P’s of Transplant Modeling to Inform Clinical Decision-making: Predictability, Probability, and Possibility

Lim, Wai H. MBBS, PhD1,2; Wong, Germaine MBBS, PhD3,4,5

Author Information
doi: 10.1097/TP.0000000000003204
  • Free

Transplantation of highly sensitized (HS) patients remains a considerable challenge for transplant programs worldwide. Approximately 5%–9% of all potential transplant candidates in the United States and Europe are HS.1 The annual transplantation rate for these patients is reduced by at least 2-fold compared with the nonsensitized individuals.2 Consequently, these patients may wait twice as long on the transplant waiting list as the unsensitized patients and are more likely to be delisted or die on the waiting list. Even with the availability of potential living donors, transplantation is not always possible among HS patients. Desensitization of a living donor/recipient pair to overcome low-level incompatibility and the deleterious donor-specific anti-HLA antibodies (DSA) is a reasonable option. However, outcomes after desensitization are less optimal and these patients are more likely to experience premature allograft failure or treatment-related complications.3

The paired kidney exchange program has been successful to match donor–recipient pairs, often without the necessity to overcome the HLA barrier through desensitization.4 Through this strategy, an improvement in the transplantation rate of HS patients has been successfully attained. In this study, de Klerk et al5 designed a computer program, the Computerised Integration of Alternative Transplantation (CIAT), which integrated all alternative living donor kidney transplantation programs in the Netherlands. The model aimed to optimize the number of transplant matches for HS patients and maximize the total number of transplants for the population of interests. The CIAT software utilizes a mathematical algorithm that effectively deals with the multiple criteria being imposed on the allocation of donors to potential transplant candidates in a paired kidney exchange system. This algorithm has been validated using kidney-exchange data from the Netherlands and the United States and appears to be effective, even for large, but realistic patient–donor pools.6

Using data from all recipients registered on the Eurotransplant waiting list and potential live-donors (including unspecified kidney donors) from Rotterdam between 2015 and 2016, the authors reported an additional 7 [of 20 (35%)] selected highly immunized (sHI) patients may be matched for transplantation using the CIAT simulation (ie, 1 sHI patient transplanted in reality versus 8 sHI patients matched in the CIAT simulation modeling). However, this improvement in transplant potential was probably not unexpected because an increase in the number of donor/patient pairs and unspecified donors available for matching would inevitably increase the probability of a potential match for a particular patient. In addition, the authors also accepted high-risk ABO-incompatible (titers below 1:512) and positive complement-dependent cytotoxicity (CDC)-crossmatch (XM) transplants in potential candidates with no compatible matches after a prespecified period of wait-time and failed match runs. Of the 8 sHI patients matched using the CIAT simulation model, 6 (75%) had moderate or high mean fluorescent intensity DSA (3 patients had DSA above 8000 mean fluorescent intensity), and 2 patients had positive CDC-XM to the CIAT-matched donor. Four (50%) patients had positive T- or B-cell flow-XM, with 3 times the channel shift thresholds for a positive test result.

However, there are important caveats, and these must be recognized when interpreting the study findings. The 3 key assumptions included in the modeling were (1) matched donor/patient pairs who were CDC or flow-XM positive or had high ABO titers received successful desensitization before proceeding to transplantation; (2) the model was unable to account for patients (sHI and non-sHI patients) who had been transplanted earlier through the standard allocation algorithm. These patients were not available for the simulation, and the downstream effects on subsequent donor kidney allocation/acceptance were not observed; and (3) clinicians must be willing to accept a chain that may require desensitization for the donor/recipient pairs with HLA incompatibility and a program that prioritized sHI patients. Such program could inadvertently reduce the probability of successful match(es) for other patients.

Nonetheless, findings from this study support a multifaceted approach that integrates a gamut of novel strategies to improve the transplant potential of HS, difficult-to-match patients. Information provided by this simulated model is also useful to inform clinical decision-making process and the trade-offs when choosing the alternatives, ensuring that the desired impact of the modification in matching policy is achieved. However, one must be cautioned about the clinical applicability of this modeling, as these findings are results of the matches and not the actual transplantations. Ultimately, the short-and long-term outcomes of these potential transplants are unknown.

REFERENCES

1. Stewart DE, Kucheryavaya AY, Klassen DK, et al. Changes in deceased donor kidney transplantation one year after KAS implementation. Am J Transplant. 2016;16:1834–1847doi:10.1111/ajt.13770
2. Heidt S, Witvliet MD, Haasnoot GW, et al. The 25th anniversary of the Eurotransplant Acceptable Mismatch program for highly sensitized patients. Transpl Immunol. 2015;33:51–57doi:10.1016/j.trim.2015.08.006
3. Keith DS, Vranic GM. Approach to the highly sensitized kidney transplant candidate. Clin J Am Soc Nephrol. 2016;11:684–693doi:10.2215/CJN.05930615
4. Roodnat JI, Kal-van Gestel JA, Zuidema W, et al. Successful expansion of the living donor pool by alternative living donation programs. Am J Transplant. 2009;9:2150–2156doi:10.1111/j.1600-6143.2009.02745.x
5. de Klerk M, Kal-van Gestel JA, Van de Wetering J, et al. Creating options for difficult-to-match kidney transplant candidates. Transplantation. 2020. doi:10.1097/TP.0000000000003203
6. Glorie KM, van de Klundert JJ, Wagelmans APM. Kidney exchange with long chains: an efficient pricing algorithm for clearing barter exchanges with branch-and-price. Manufactur Service Operat Management. 16:481–603doi:10.1287/msom.2014.0496
Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.