Pancreas

OPTN/SRTR 2017 Annual Data Report: Pancreas

Abstract

In 2017, 1492 patients were added to the pancreas transplant waiting list, 964 listed as active, a slight increase from 2016. This is significant because for the first time in the past decade, the steady downward trend in additions to the waiting list has been reversed. Proportions of pancreas donors with cerebrovascular accident as cause of death decreased, with a concomitant increase in proportions with anoxia and head trauma. This is partly a result of the national opioid crisis, and it reflects increasing use of younger donors for pancreas transplant. The 2017 outcome report remains compromised by previous variation in reporting graft failure. Although the OPTN Pancreas Transplantation Committee has approved more precise definitions of pancreas graft failure, implementation of these definitions took place recently, and the data are not reflected in this report.

Introduction

Below is a summary of recent activity of the OPTN/UNOS Pancreas Transplantation Committee:

A pancreas after kidney (PAK) guidance paper passed by the OPTN/UNOS Board in December 2017 provides guidance and data on how PAK may be an underused option that should be considered more frequently than it is. (See Fridell JA et al. The survival advantage of pancreas after kidney transplant. Am J Transplant. 2018; doi: 10.1111/ajt.15106. [Epub ahead of print]).
Simultaneous kidney-pancreas (SPK) waiting time policy was changed in a policy passed by the OPTN/UNOS Board in June 2018. Waiting time will no longer require that candidates with a C-peptide >2 have a body mass index (BMI) below the maximum. This change is set to be implemented in 2019.
A proposal updating islet bylaws was on consent at the regional meetings during fall 2018 public comment and was approved with amendments by the OPTN/UNOS Board in December 2018. The proposed changes to the islet Bylaws will better reflect the needs of islet programs and the expertise necessary to ensure patient safety.
A proposal changing pancreas program functional inactivity was generally supported during fall 2018 public comment and was approved by the OPTN/UNOS Board in December 2018. This project improves the pancreas program functional inactivity definition to provide more information to patients and avoid patients at low-volume transplant programs languishing on the waiting list.
The OPTN/UNOS Pancreas Committee is working with the OPTN/UNOS Kidney Committee to remove and replace donation service area and region as geographic units in kidney and pancreas allocation. This proposal will go out for public comment in spring 2019.

Waiting List

In 2017, 1492 patients were added to the pancreas transplant waiting list, 964 listed as active, a slight increase from 2016. This is significant because for the first time in the past decade, the steady downward trend in additions to the waiting list has been reversed. Increases were noted for SPK and pancreas transplant alone (PTA) listings, but there were continued decreases for PAK listings (Figure PA 1, Figure PA 2). Despite the number of additions to the waiting list outpacing the number of transplants, total prevalent candidates on the list continued a downward trend, perhaps accounted for by death on the waiting list or removal for other reasons (Figure PA 3, Figure PA 4). This trend was consistent across all transplant types.

The proportion of older candidates (aged ≥50 years) continued to increase slightly and comprised 26.6% of the waiting list in 2017 (Figure PA 5). This has been attributed to an increase in listings of candidates with type II diabetes, and the proportion of these candidates on the waiting list increased slightly to 11.7% in 2017 (9.8% in 2016) (Figure PA 8). This number is important because it falls within the 10%-15% range originally proposed by the OPTN/UNOS Pancreas Transplantation Committee as the acceptable proportion of candidates with type II diabetes on the waiting list while setting BMI limits for waiting time accrual for this group.

Gender distribution on the waiting list for 2017 was 55.4% male and 44.6% female (Figure PA 6), the first time in several years that the proportion of men increased. The proportion of white candidates continued a downward trend, at the lowest in over a decade (59.5%) in 2017, with a concomitant increase in black, Hispanic, and Asian candidates (Figure PA 7). BMI distribution among waitlisted candidates remained largely unchanged, with about 40% in the range of 18.5-25 kg/m² in 2017 (Figure PA 10). SPK candidates accounted for 73.2% of the list, PTA candidates 14.8%, and PAK candidates 12.1% (Figure PA 11). As noted, proportions of PAK transplants continued decreasing while SPK and PTA increased gradually.

The proportion of prevalent candidates waiting <1 year for transplant increased to 45.4% in 2017 from 41.8% in 2016, presumably due to more new additions. However, the proportion waiting ≥1 year (1-2 and 2-3 years) decreased, suggesting that candidates were actively undergoing transplant during this time period (Figure PA 9).

Deceased donor transplant rates among candidates with type I diabetes continued to increase over the past 2 years and reached 39.4 transplants per 100 waitlist-years in 2017; for candidates with type II diabetes, the rate remained flat in 2017 (45.4), after an increase in 2016 (compared with 2015) (Figure PA 12). This may represent stabilization of transplant rates in candidates with type II diabetes after the new pancreas allocation system was implemented in late 2014. Transplant rates for SPK increased to 47.4 per 100 waitlist-years in 2017 from 42.6 in 2016, and rates for PTA remained flat. Interestingly, rates for PAK increased to 18.2 in 2017 from 14.3 in 2016, presumably from contraction of the waiting list considering that the number of PAK transplants declined (Figure PA 13).

Among candidates listed in 2014, 55.2% of SPK candidates, 50.6% of PTA candidates, and only 39.7% of PAK candidates underwent deceased donor transplant within 3 years (Figure PA 16, Figure PA 15, Figure PA 14), and 27%-34% of waiting pancreas candidates (PTA and PAK, respectively) were removed from the list without transplant or death during that period. Median months to transplant in 2016-2017 decreased to 12.9 for SPK and 22.8 for PTA, and could not be calculated for PAK because less than 50% of the cohort had undergone transplant, as has been the case since 2010 (Figure PA 17). No consistent pattern emerged in geographic distribution among the transplant types regarding percentage of candidates undergoing transplant within 2 years of listing, aside from a very high degree of disparity (Figure PA 18, Figure PA 19, Figure PA 20). A range of 0%-100% of candidates underwent transplant within each transplant type.

Overall pretransplant mortality rates decreased consistently over the years to 5.9 per 100 waitlist-years in 2017, compared with 7.5 in 2006 (Figure PA 23). As expected, mortality was highest for SPK candidates at 7.2 per 100 waitlist-years, compared with 5.40 for PAK and 3.1 for PTA. Waitlist mortality did not differ substantially by race (6.6 per 100 waitlist-years for black versus 6.2 for white candidates), although rates among the non-white/non-black race category have been consistently lower over the last 4 years (Figure PA 22). Of note, waitlist mortality declined in 2017 among candidates aged 35-49 years from 6.8 to 5.1 per 100 waitlist-years (Figure PA 21).

Pretransplant mortality rates by donation service area for the 2016-2017 cohort demonstrated geographic differences ranging from 0 to 22.2 per 100 waitlist-years (Figure PA 24). Death within 6 months after waitlist removal occurred for 5.8% of SPK, 4.3% of PAK, and 0% of PTA candidates (Figure PA 25). No deaths occurred among candidates aged >50 years (Figure PA 26).

Donations

No living donor pancreas transplants were reported in the United States in the past few years. Most deceased pancreas donors were aged <35 years, but the proportion of donors aged 35-64 years increased marginally in 2017 (Figure PA 27, Figure PA 28). No reported pancreas donors were aged ≥65 years. Most donors were white (63.8% in 2017); black and Hispanic donors account for 33.1% (Fig. 29).

The overall discard rate for pancreata recovered for transplant remained high and increased with donor age. Only 12.6% of pancreata were discarded from donors aged <18 years, compared with 72.7% from donors aged ≥50 years (Figure PA 30). Discard rates were unchanged for white donors at 24.6%, and declined for black donors from 23.2% in 2016 to 16.7% in 2017. Whether this is a function of the age demographic or other factors should be studied (Figure PA 31). Not surprisingly, the discard rate was directly correlated with donor BMI, and increased with increasing BMI (Figure PA 32). Over 80% of pancreata from donors with BMI ≥35 kg/m² were discarded in 2017. Encouragingly, in the past 3 years, use of pancreata from Public Health Service increased-risk donors did not differ appreciably from use of other pancreata. Increased sensitivity of screening tests for transmissible infections contributed to this (Figure PA 33).

Average calculated pancreas donor risk index (PDRI) decreased steadily over the past decade. Overall PDRI was 1.04 in 2017, 0.98 for PAK, 1.11 for PTA, and 1.04 for SPK (Figure PA 35). Considering components of the PDRI, the proportion of donors with cerebrovascular accident (CVA) as cause of death declined remarkably to 7.9% in 2017. Use of donation after circulatory death donors remained low (<3%) (Figure PA 34). With the proportion of CVA decreasing as a cause of donor death came an expected increase in proportions of anoxia and head trauma. This is partly a result of the national opioid crisis, and it reflects increasing use of younger donors for pancreas transplant (Figure PA 36).

Transplants

The overall number of pancreas transplants remained stable, with 1002 transplants performed in 2017 (1013 in 2016). Of these, 213 were solitary pancreas transplants (PAK or PTA) (Figure PA 37). Enthusiasm for performing SPK transplants continued, especially for ages 35-49 years (Figure PA 38). The percentage of black transplant recipients has tended to increase over the past 5 years (Figure PA 40). Surprisingly, the proportion of type II diabetic recipients was unchanged from 2016, at about 10% of overall transplants (Figure PA 41). The concern that proportions of recipients with type II diabetes might increase substantially with relaxation of the upper limit of BMI from 28 to 30 has been somewhat allayed, at least for now.

Induction therapy in pancreas transplantation is predominantly based on T-cell depletion. The proportion of recipients receiving T-cell depleting agents has consistently increased over the past decade and is currently at its highest level; almost 87% of recipients received T-cell depletion therapy in 2017 (Figure PA 43). Of note, an uptick in steroid avoidance occurred in the context of Tac/MMF maintenance, with 40% of recipients receiving Tac/MMF with no steroids (Figure PA 44).

The proportion of unsensitized recipients ranged from approximately 60%-70%, depending on transplant type. No clear trends over time appeared, except for SPK, where the proportion of unsensitized recipients seemed to shrink over recent years (Figure PA 45, Figure PA 46, Figure PA 47). More PAK recipients were highly sensitized (c/PRA 80%-100%) than either SPK or PTA recipients. The overwhelming majority of pancreas transplant recipients (>75%) had >3 HLA mismatches. Despite some reports showing the immunologic value of -DR matching in SPK transplants and B locus matching in solitary pancreas transplants, there is little enthusiasm for overall HLA matching (Figure PA 48).

Regarding program volume, most pancreas transplants (about two-thirds) are performed at programs performing >10 pancreas transplants per year. In this group, the proportion of high-volume programs increased (>25 annually) over recent years. Not surprisingly, very low-volume programs (1-2 transplants per year) accounted for <5% of transplants (Figure PA 50).

Outcomes

The 2017 outcome report remains compromised by previous variation in reporting graft failure. Although the OPTN/UNOS Pancreas Transplantation Committee has approved more precise definitions of pancreas graft failure, implementation of these definitions took place in early 2018, and the data are not reflected in this report. Some of the definitions are concrete, such as: 1) a recipient’s transplanted pancreas is removed; 2) a recipient re-registers for pancreas transplant; 3) a recipient registers for an islet transplant after undergoing pancreas transplant; or 4) a recipient dies. Pancreas graft failure can also be defined if a recipient’s total insulin use is greater than or equal to 0.5 units/kg/day for a consecutive 90 days. The latter definition may be problematic if the recipient’s starting insulin dose is less than 0.5 units/kg/day. Nonetheless, these new definitions are a step in the right direction, and will provide an opportunity to analyze graft failure, based on a uniform definition, in subsequent SRTR reports. Although this report does not reflect the newer definitions, program-reported early pancreas graft failure has remained relative stable over the past 3 years, with rates of 9.2%, 7.3%, and 7.1 % for PAK, PTA, and SPK, respectively (Figure PA 51).

Unlike the variability in the data reported to OPTN/UNOS for pancreas graft outcomes, the data for kidney transplant graft outcomes following SPK and PAK are well defined. All-cause kidney graft failure following SPK was 4.9%, 16.6%, and 34.9% at 1, 5 and 10 years, respectively (Figure PA 52). These excellent results remained superior to results for non-SPK deceased donor kidney transplants, reflecting the higher quality of deceased donor kidneys used in SPK. All-cause kidney graft failure following deceased donor kidney PAK was 2.9%, 16.1%, and 47.1% at 1, 5, and 10 years, respectively (Figure PA 54). The slightly poorer long-term outcomes for PAK were likely related to kidney outcomes for PAK being based on time from pancreas transplant. The 10-year death-censored kidney graft failure rates for SPK and PAK were 21.4% (Figure PA 53) and 21.2% (Figure PA 55), respectively, and these high rates of graft survival demonstrated the high quality of kidneys (low KDPI) used in both SPK and PAK.

The data on patient survival following pancreas transplant (SPK, PAK, PTA) show nearly 18,500 pancreas recipients alive in 2017 (Figure PA 58). Patient mortality continued to decrease for all pancreas transplant categories, with 5-year mortality for PAK, PTA, and SPK at 8.1 %, 11.7%, and 8.8%, respectively (Figure 65). The long-term mortality data reflect the cardiovascular comorbidity in this population, with 10-year mortality of 28.9%, 20.7%, and 24.6% for PAK, PTA, and SPK, respectively (Figure PA 66). One-year mortality for PTA transplants in 2015-2016 was 0% (Figure PA 64), demonstrating improvements in patient selection, surgical technique, and immunosuppression management in these challenging recipients. Interestingly, 5-year patient survival following pancreas transplant was similar for transplants in recipients with type 1 (91%) and type 2 diabetes (93%) (Figure PA 68). Although patients with type 2 diabetes are usually older with more cardiovascular risk factors, similar mortality between the two groups speaks to good judgement in patient selection and preoperative assessment of cardiovascular risks. It will be important to gain a better understanding of the long-term metabolic benefits following pancreas transplant in recipients with type II diabetes.

Incidence of a first rejection episode remained consistently low for all categories of pancreas transplant for transplants performed in 2015-2016, and were 16.2%, 16.3%, and 13.7% following PAK, PTA, and SPK, respectively (Figure PA 59). These extremely low rejection rates clearly reflect ongoing improvements in immunosuppression protocols. Importantly, the cumulative incidence of posttransplant lymphoproliferative disorder is quite high in EBV-naive PTA recipients (6.1%), compared with 2.9% and 1.3% in EBV-naive SPK and PAK recipients, respectively (Figure PA 61, Figure PA 62, Figure PA 63). This is likely related to rigorous immunosuppressive regimens used to prevent rejection in the immunologically-challenging PTA recipients.

Figure List

Waiting list

Deceased donation

Figure PA 27. Deceased pancreas donor count by age
Figure PA 28. Distribution of deceased pancreas donors by age
Figure PA 29. Distribution of deceased pancreas donors by race
Figure PA 30. Rates of pancreata recovered for transplant and not transplanted by donor age
Figure PA 31. Rates of pancreata recovered for transplant and not transplanted by donor race
Figure PA 32. Rates of pancreata recovered for transplant and not transplanted by donor BMI
Figure PA 33. Rates of pancreass recovered for transplant and not transplanted, by donor risk of disease transmission
Figure PA 34. Donor-specific components of the pancreas donor risk index
Figure PA 35. Average pancreas donor risk index
Figure PA 36. Cause of death among deceased pancreas donors

Transplant

Figure PA 37. Total pancreas transplants
Figure PA 38. Total pancreas transplants by age
Figure PA 39. Total pancreas transplants by sex
Figure PA 40. Total pancreas transplants by race
Figure PA 41. Total pancreas transplants by diagnosis
Figure PA 42. Total pancreas transplants by body mass index (BMI)
Figure PA 43. Induction agent use in adult pancreas transplant recipients
Figure PA 44. Immunosuppression regimen use in adult pancreas transplant recipients
Figure PA 45. C/PRA at time of transplant in adult recipients of pancreas after kidney transplant
Figure PA 46. C/PRA at time of transplant in adult recipients of pancreas transplant alone
Figure PA 47. C/PRA at time of transplant in adult recipients of simultaneous kidney-pancreas transplant
Figure PA 48. Total HLA A, B, and DR mismatches among adult pancreas transplant recipients, 2013-2017
Figure PA 49. Annual adult pancreas transplant center volumes, by percentile
Figure PA 50. Distribution of adult pancreas transplants by annual center volume

Outcomes

Table List

Waiting list

Table PA 1. Demographic characteristics of adults on the pancreas transplant waiting list on December 31, 2017
Table PA 2. Clinical characteristics of adults on the pancreas transplant waiting list on December 31, 2017
Table PA 3. Listing characteristics of adults on the pancreas transplant waiting list on December 31, 2017
Table PA 4. Transplant waitlist activity among adults waiting for a pancreas after kidney transplant
Table PA 5. Transplant waitlist activity among adults waiting for a pancreas transplant alone
Table PA 6. Transplant waitlist activity among adults waiting for a simultaneous kidney pancreas transplant
Table PA 7. Removal reason among adults waiting for pancreas after kidney transplant
Table PA 8. Removal reason among adults waiting for pancreas transplant alone
Table PA 9. Removal reason among adults waiting for simultaneous kidney-pancreas transplant

Transplant

Table PA 10. Demographic characteristics of adult pancreas transplant recipients, 2017
Table PA 11. Clinical characteristics of adult pancreas transplant recipients, 2017
Table PA 12. Transplant characteristics of adult pancreas transplant recipients, 2017
Table PA 13. Adult pancreas donor-recipient serology matching, 2013-2017

A line plot for new adult candidates added to the pancreas transplant waiting list; the pak category decreases by 74.5% from 447 candidates at 2006 to 114 candidates at 2017; the pta category decreases by 27.7% from 292 candidates at 2006 to 211 candidates at 2017; the spk category decreases by 24.3% from 1541 candidates at 2006 to 1167 candidates at 2017; and the all category decreases by 34.6% from 2280 candidates at 2006 to 1492 candidates at 2017. — **Figure PA 2. New adult candidates added to the pancreas transplant waiting list**
A new candidate is one who first joined the list during the given year, without having been listed in a previous year. Previously listed candidates who underwent transplant and subsequently relisted are considered new. Candidates concurrently listed at multiple centers are counted once. PAK, pancreas after kidney; PTA, pancreas transplant alone; SPK, simultaneous kidney pancreas transplant.

A line plot for adults actively listed for pancreas transplant on december 31 each year; the pak category decreases by 81.4% from 291 candidates at 2006 to 54 candidates at 2017; the pta category decreases by 37.5% from 184 candidates at 2006 to 115 candidates at 2017; the spk category decreases by 40.6% from 1442 candidates at 2006 to 856 candidates at 2017; and the all category decreases by 46.5% from 1917 candidates at 2006 to 1025 candidates at 2017. — **Figure PA 3. Adults actively listed for pancreas transplant on December 31 each year**
Candidates concurrently listed at multiple centers are counted once. Those with concurrent listings and active at any program are considered active. PAK, pancreas after kidney; PTA, pancreas transplant alone; SPK, simultaneous kidney pancreas transplant.

A line plot for adults listed for pancreas transplant on december 31 each year; the pak category decreases by 66.7% from 986 candidates at 2006 to 328 candidates at 2017; the pta category decreases by 30.1% from 528 candidates at 2006 to 369 candidates at 2017; the spk category decreases by 28.0% from 2298 candidates at 2006 to 1655 candidates at 2017; and the all category decreases by 38.3% from 3812 candidates at 2006 to 2352 candidates at 2017. — **Figure PA 4. Adults listed for pancreas transplant on December 31 each year**
Candidates concurrently listed at multiple centers are counted once. PAK, pancreas after kidney; PTA, pancreas transplant alone; SPK, simultaneous kidney pancreas transplant.

A line plot for distribution of adults waiting for pancreas transplant by age; the 18 to 34 category is 19.6 percent at 2006 and remains relatively constant with a value of 21.4 percent at 2017; the 35 to 49 category decreases by 11.0% from 58.4 percent at 2006 to 52 percent at 2017; and the greater than or equal to 50 category increases by 21.1% from 22 percent at 2006 to 26.6 percent at 2017. — **Figure PA 5. Distribution of adults waiting for pancreas transplant by age**
Candidates waiting for transplant at any time in the given year. Candidates listed concurrently at multiple centers are counted once. Age is determined at the later of listing date or January 1 of the given year. Active and inactive candidates are included.