Successful Isolated Intestinal Transplantation in Sensitized Recipients With the Use of Virtual Crossmatching

Authors


Abstract

We evaluated virtual crossmatching (VXM) for organ allocation and immunologic risk reduction in sensitized isolated intestinal transplantation recipients. All isolated intestine transplants performed at our institution from 2008 to 2011 were included in this study. Allograft allocation in sensitized recipients was based on the results of a VXM, in which the donor-specific antibody (DSA) was prospectively evaluated with the use of single-antigen assays. A total of 42 isolated intestine transplants (13 pediatric and 29 adult) were performed during this time period, with a median follow-up of 20 months (6–40 months). A sensitized (PRA ≥ 20%) group (n = 15) was compared to a control (PRA < 20%) group (n = 27) to evaluate the efficacy of VXM. With the use of VXM, 80% (12/15) of the sensitized patients were transplanted with a negative or weakly positive flow-cytometry crossmatch and 86.7% (13/15) with zero or only low-titer (≤1:16) DSA. Outcomes were comparable between sensitized and control recipients, including 1-year freedom from rejection (53.3% and 66.7% respectively, p = 0.367), 1-year patient survival (73.3% and 88.9% respectively, p = 0.197) and 1-year graft survival (66.7% and 85.2% respectively, p = 0.167). In conclusion, a VXM strategy to optimize organ allocation enables sensitized patients to successfully undergo isolated intestinal transplantation with acceptable short-term outcomes.

Abbreviations: 
CDC-XM

complement-dependent cytotoxic crossmatch

CIT

cold ischemia time

CMV

cytomegalovirus

DSA

donor specific antibody

FC-XM

flow cytometry crossmatch

ITx

intestine transplantation

MCS

median channel shift

MFI

mean fluorescence intensity

NOD

nucleotide-binding oligomerization domain-containing protein 2

PRA

panel reactive antibody (%)

PTLD

posttransplant lymphoproliferative disorder

SGS

short gut syndrome

TPN

total parenteral nutrition

VXM

virtual crossmatch

Introduction

Transplantation in the sensitized, isolated intestine recipient remains a significant challenge for the intestine transplant community. In the setting of an already highly immunogenic transplant, sensitization has been shown to increase the incidence and severity of rejection following intestinal transplantation. In particular, the sensitized isolated intestine recipient with a positive complement-dependent cytotoxic crossmatch (CDC-XM) has a high incidence of early treatment-refractory rejection and graft loss (1,2). Based on these early experiences, a positive CDC-XM is generally considered prohibitive for an isolated intestine transplant. Consequently, transplanting the sensitized patient is extraordinarily difficult due to the increased risk of a positive CDC-XM added to a shortage of high-quality intestine donors.

Unfortunately, many patients who require intestine transplantation are sensitized due to a history of multiple transfusions, line infections, pregnancy and in some cases retransplantation (3–5). The predominant risk in sensitized patients is the formation of donor-specific antibody (DSA). With the evolution of solid phase assays, the ability to detect and minimize DSA is possible. Single-antigen detection of DSA has led to the recent application of the virtual crossmatch (VXM), in which known recipient human leukocyte antigen (HLA)-specific antibodies are prospectively compared to donor HLA type. The VXM has been shown to accelerate kidney and heart allocation in the disadvantaged sensitized patient, and a negative VXM was associated with a very low risk for early rejection and good allograft survival (6–13). The introduction of the VXM has also provided an opportunity to transplant the immuno-logically high-risk sensitized isolated intestine recipient.

The VXM is ideally suited for improving organ allocation in the sensitized isolated intestine transplant recipient due to a number of reasons. First, with the limited supply of high-quality intestine donors and a nationally based allocation system, the majority of intestine donors are often obtained from centers at considerable distance from the originating institution. Without the use of VXM, considerable logistical resources must be utilized in order to perform a CDC-XM, as physical samples of both donor and recipient serum must be present and available for testing. This dictated that the CDC-XM be performed at the time of arrival of the intestine organ to our institution. As with any highly sensitized recipient, positive CDC-XM and transplant cancellations were an often occurrence. With the use of VXM, the risk of a positive CDC-XM is negligible and logistical resources can be utilized in a cost-effective manner to obtain an intestine allograft from a distant region. Secondly, the VXM obviates the need to wait for a conventional CDC-XM, thus minimizing cold ischemia time (CIT) for the intestine graft. Finally, the VXM can be used to reduce a recipient's immunologic risk based on DSA, and immunosuppression management can be individualized. At our institution, highly sensitized isolated intestine candidates were disadvantaged by our requirement for a negative CDC-XM prior to performing an isolated intestine transplant. Because of this and the growing number of highly sensitized isolated intestine candidates on our list, a VXM protocol was initiated in September 2008.

In this study, we present our experience utilizing VXM to optimize organ allocation and minimize immunologic risk in sensitized recipients of isolated intestine transplants. All sensitized recipients underwent VXM for organ allocation without the requirement for a CDC-XM. We compared patient and graft outcomes in sensitized (PRA ≥ 20%) to control (PRA < 20%) recipients to determine the efficacy of the VXM. To our knowledge, this is the first application of a VXM strategy toward intestine transplantation.

Materials and Methods

Study population

All patients who underwent isolated intestinal transplantation at our institution following the implementation of the VXM (after September 2008 until June 2011) were included in this study. All transplant recipients achieved standard criteria for intestine transplantation. Donors were selected based on ABO matching, age, size, history of intestinal disease and cytomegalovirus (CMV) status. Patients were considered sensitized with PRA ≥ 20% and were compared to a control group of patients with PRA < 20%. Primary endpoints were 1-year freedom from rejection, patient and graft survival. A prospectively maintained database was used for analysis. This study was approved by the Institutional Review Board.

Antibody detection and specificity analysis

HLA class I and class II antibody testing was done using either the FlowPRA® Single Antigen or the LABScreen® Single Antigen assay (One Lambda, Canoga Park, CA, USA) according to the manufacturer's protocol. For the LABScreen® Single Antigen assay, specificities were called positive when the mean fluorescence intensity (MFI) was greater than 2,000. Dilution studies were done to determine the titer of all antibody specificities detected. Patient serum was monitored for the presence of HLA specific antibody at monthly intervals to establish baseline antibody specificity information and titers. Titration studies were done at the time of a patient's initial evaluation, when new specificities were detected and when a significant change in MFI was observed.

Crossmatching

Beginning in September 2008, allograft allocation in sensitized recipients was based on the results of a VXM, in which the donor-specific antibody (DSA) was evaluated. At the time of a deceased donor organ offer, the titer, MFI and total number of DSA were considered for virtual crossmatching. DSA titers greater than 1:16 were generally considered prohibitive for isolated ITx. Prior to the use of VXM, a complement-dependent cytotoxic crossmatch (CDC-XM) was always performed before intestine transplantation and potential recipients were excluded when the crossmatch was positive. With implementation of the VXM, the requirement for a pretransplant CDC-XM was no longer mandatory prior to transplantation.

For all transplant recipients, a T cell and B cell flow cytometry crossmatching (FC-XM) was performed at time of transplant, but the results were not available prior to transplant. A three-color flow cytometric crossmatch (FC-XM) was performed according to standard techniques (14). T cell and B cell FC-XMs were performed using pretransplant recipient sera drawn on the day of transplant and at least one historic sera acquired within the previous 12 months along with donor lymphocytes. Interpretation of results was based on the median fluorescence of the patient sera tested with donor lymphocytes compared to the median fluorescence of the negative control sera tested with donor lymphocytes. T cell FC-XMs were considered to be positive when the median channel shift (MCS) was greater than 50. B cell FC-XMs were considered to be positive when the MCS was greater than 100. Results were considered to be weakly positive when the MCS was within 10–20 MCS of cutoff for a T cell FC-XM and within 20–30 MCS for a B cell FC-XM.

Immunosuppression and rejection grading

Standard immunosuppression (IS) consisted of IL2 blockade (basiliximab) induction with maintenance IS of tacrolimus, sirolimus and prednisone as previously described (15). Sensitized recipients generally received thymoglobulin (1.5 mg/kg/day ×5 days) induction (n = 9) and intravenous immunoglobulin (IVIG, 2 gm/kg/day ×1 day) (n = 9) followed by standard maintenance IS (16). There was some variation in this practice during the study period as our experience with the VXM in sensitized recipients evolved. Currently, the use of thymoglobulin and IVIG is now reserved only for sensitized recipients with a positive FC-XM and/or DSA. As a result, some of the sensitized recipients (n = 6) received standard IS with basiliximab induction. Intestinal biopsies were obtained on the protocol or based on clinical symptoms. Standard grading was used for histological diagnosis of acute rejection (AR) (17).

Statistics

Continuous variables were summarized by mean ± standard deviation. Normally distributed continuous variables were compared between sensitized and control groups by the independent sample t-test. Nonnormally distributed variables were compared across groups by the Wilcoxon rank sum test. Categoric variables were summarized by frequency and percent. Categoric variables were compared across groups by the Fisher exact test. Survival and freedom from rejection were estimated using the Kaplan–Meier method and were compared across groups by the log-rank test. All significance tests were two sided. A p value less than 0.05 was considered significant. The SPSS version 16.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis.

Results

Fifty-eight intestinal transplants were performed at our institution from September 2008 to June 2011, including 42 isolated intestine transplants (excluding multivisceral, modified multivisceral and liver/intestine transplants) in 42 patients. There were 13 pediatric and 29 adult isolated intestine transplant recipients with a median follow-up of 20 (range 6–40) months. Four retransplants were performed; all were referred from other institutions or received their primary transplant at our institution prior to 2008. There were 15 sensitized (PRA ≥ 20%) and 27 control (PRA < 20%) recipients.

Donor and recipient clinical characteristics

Donor and recipient demographic and clinical characteristics were generally well matched between sensitized and control groups, including recipient gender, age, colon or kidney graft inclusion, cold ischemia time, surgery case time, donor/recipient age and weight ratio, HLA mismatch, cytomegalovirus (CMV) high risk (donor positive/recipient negative) and recipient nucleotide-binding oligomerization domain-containing protein 2 (NOD2) genotype status (which has previously been shown to be associated with rejection in intestinal transplantation) (18) (Table 1). There was a significantly higher incidence of short gut syndrome in the sensitized group. There were more retransplants in the sensitized group but this did not reach statistical significance. The follow-up was significantly longer in the control group.

Table 1. Donor and recipient demographic and clinical characteristics
 Control (n = 27)Sensitized (n = 15)p-Value
  1. 1Mean ± SD.

  2. 2One recipient did not have NOD2 genotyping.

Recipient gender-–no. (%)  0.202
 Male17 (63)6 (40) 
 Female10 (37)9 (60) 
Recipient age-–no. (%)  0.739
 ≤18 years9 (33.3)4 (26.7) 
 >18 years18 (66.7)11 (73.3) 
Primary pathology-–no. (%)
 Short gut syndrome7 (25.9)11 (73.3) 0.004
 Mesenteric thrombosis9 (33.3)1 (6.7)0.068
 Dysmotility disorder5 (18.5)1 (6.7)0.395
 Neoplastic disorder2 (7.4)00.530
 Other4 (14.8)2 (13.3)1.000
Retransplant-–no. (%)1 (3.7)3 (20)0.122
Allograft implanted-–no. (%)
 Small bowel and colon16 (59.3)8 (53.3)0.754
 Small bowel and kidney2 (7.4)1 (6.7)1.000
Surgery case time16:14 ± 1:366:25 ± 1:260.730
Donor/recipient age ratio10.55 ± 0.560.51 ± 0.230.784
Donor/recipient weight ratio10.77 ± 0.290.80 ± 0.190.749
HLA A, B, DR mismatch14.9 ± 0.85.1 +0.80.402
CMV high risk-–no. (%)1 (3.7)01.000
Recipient NOD2 status-–no. (%)2  0.636
 Wild type22 (81.5)14 (93.3) 
 Mutant4 (14.8)1 (6.7) 
Follow-up (months)124.1 ± 11.715.4 ± 9.2 0.018

Crossmatch and donor-specific antibody

PRA, crossmatch and DSA data were compared between the sensitized and control groups (Table 2). There were three patients with a positive FC-XM in the sensitized group, including one patient with both positive B cell and T cell FC-XM and two patients with a positive B cell FC-XM. There were eight patients in the sensitized group with DSA, three with class I DSA and five with class II DSA. There were five additional patients with weakly positive B cell FC-XM in the sensitized group, four of which had low titer DSA (≤1:16). There was also one patient in the control group with a weakly positive B cell FC-XM who also had class II DSA at time of transplant. The entire sensitized cohort is depicted in detail in Table 3.

Table 2. Crossmatch and DSA data
 Control (n = 27)Sensitized (n = 15)p-Value
  1. 1Mean ± SD.

Panel reactive
antibody1
 Class I 1.5 ± 3.6 33.5 ± 33.8 0.003
 Class II0.6 ± 240.7 ± 38 0.001
Positive crossmatch—no. (%)
 Positive T cell FC-XM01 (6.7)0.357
 Positive B cell FC-XM03 (20) 0.040
 Positive T & B cell FC-XM01 (6.7)0.357
 Any positive FC-XM03 (20) 0.040
Pretransplant DSA—no. (%)
 Class I DSA03 (20) 0.040
 Class II DSA1 (3.7)5 (33.3) 0.016
 Class I & II DSA00
 Any DSA1 (3.7)8 (53.3) <0.001  
Table 3. Sensitized group clinical characteristics
PtAge/SexDiagnosisPRA Class IPRA Class IIT cell FC-XMB cell FC-XMPretransplant DSARejectionSurvival (follow-up)
  1. AR = acute rejection; cell FC-XM = cell flow cytometry crossmatch; DSA = donor specific antibody; Gr = grade; PRA = panel reactive antibody (%); PTLD = posttransplant lymphoproliferative disorder; Rx = treatment; SBO = small bowel obstruction; SGS = short gut syndrome; cell FC-XM = cell flow cytometry crossmatch; thymo = thymoglobulin.

 146FSGS (SBO)9499Neg Weak Pos DR4 1:16 DR8 1:16Gr3 AR day 107 (Rx thymo), refractive, explant day 170Dead, rejection (day 217)
 222MMesenteric thrombosis25 0NegNegNegNoneDead, suicide (day 733)
 346MRetransplant (mesenteric thrombosis)2299Neg Pos DQ5 1:128Gr3 AR day 93 (Rx thymo), chronic rejection, explant day 405Dead, sepsis (day 473)
 442FPseudo-obstruction 056Neg Pos DR17 1:64Gr2 AR day 243 (Rx thymo, alemtuzumab) Gr3 AR day 366 (Rx thymo, infliximab)Dead, rejection (day 511)
 57MSGS (intestinal atresia)78 0NegNegB7 1:4 B44 1:8 A25 1:8NoneAlive, functioning graft (day 1063)
 63MSGS (volvulus) 025NegNegNegNoneAlive, functioning graft (day 494)
 761FSGS (SBO)3167Neg Weak Pos NegNoneDead, sepsis (day 107)
 825FCongenital secretory diarrhea 057NegNegNegNoneDead, sepsis (day 365)
 941FSGS (SBO)84 0 Pos Pos B35 1:16NoneAlive, functioning graft (day 611)
1010FRetransplant (tufting enteropathy)1796Neg Weak Pos DQ7 1:16NoneDead, sepsis (day 134)
1153MSGS (SBO) 057Neg Weak Pos DR4 1:8Gr1 AR day 19 (Rx steroids), Gr1 AR day 26 (Rx thymo), Gr3 day 211 (Rx explant)Dead, PTLD (day 222)
1224FSGS (SBO)6628Neg Weak Pos B27 1:8Gr2 AR day 153 (Rx thymo)Alive, functioning graft (day 431)
1353MRetransplant (trauma) 027NegNegNegGr1 AR day 15 (Rx thymo), Gr2 AR day 33 (Rx steroids)Alive, functioning graft (day 423)
142FSGS (necrotizing enterocolitis)27 0NegNegNegNoneAlive, functioning graft (day 288)
1536FSGS (SBO)59 0NegNegNegGr2 AR day 120 (Rx steroids)Alive, functioning graft (day 255)

All DSA titers were 1:16 or less, with the exception of two patients. Both patients were transplanted early in our experience with VXM. One patient was highly sensitized and underwent retransplantation with a high titer DSA (1:128) and a positive B cell FC-XM in attempt to reverse impending liver failure due to intestine failure-associated liver disease. Unfortunately, he suffered from severe rejection and later required explant for chronic rejection, eventually succumbing to sepsis. The other patient was transplanted with a DSA titer of 1:64 and also later developed refractive rejection which ultimately resulted in death. After these cases no additional patients were transplanted with a DSA >1:16.

In summary, with the use of VXM in the sensitized isolated intestine recipients, 80% (12/15) were transplanted with a negative or weakly positive FC-XM and 86.7% (13/15) with zero or only low-titer DSA. The VXM appeared to correlate well with the FC-XM, as 87.5% (7/8) of patients with a positive or weakly positive FC-XM had DSA prospectively revealed by the VXM. In one case, the VXM was more sensitive than the FC-XM as one patient with DSA had a negative FC-XM.

Wait-list and cold ischemia time

Sensitized patients have longer wait times for organs relative to nonsensitized patients, primarily due to the risk of a positive CDC-XM (19). With the use of VXM, where the risk of a positive crossmatch is negligible, the mean wait-list time was not significantly different for sensitized compared to control recipients, 307 ± 327 and 200 ± 294 days respectively, p = 0.314 (Figure 1A). In addition, the majority of high-quality intestine allografts are procured from centers at considerable distance from our institution, resulting in increased the cold ischemia time (CIT) of the intestine allograft. Prior to the VXM, the requirement for a conventional crossmatch result added to the already critical CIT, increasing the difficulty of transplanting a sensitized recipient. With the VXM, sensitized recipients were able to receive organs from distant regions, and the distance traveled to a donor was not significantly different for sensitized compared to control recipients, 642 ± 503 and 655 ± 485 miles respectively, p = 0.936 (Figure 1B). In addition, the CIT was not significantly different for sensitized compared to control recipients, 7:02 ± 1:47 and 7:05 ± 1:59 hours respectively, p = 0.945 (Figure 1C).

Figure 1.

(A) Wait list time (days), (B) donor distance from Washington DC (miles) or (C) cold ischemia time (hours) in sensitized compared to nonsensitized recipients. Scatter plots with bar indicating mean, p-value by t-test.

Acute rejection

Rates of rejection were compared between the sensitized and control groups (Table 4). No statistical difference was observed in the overall incidence of acute rejection, number of acute rejection episodes, rejection grade, time to first rejection, incidence of chronic rejection, or of specific rejection treatments. The duration of rejection episodes, use of thymoglobulin and incidence of explantation was increased in the sensitized group, but this did not reach statistical significance. There was no significant difference in 1-year freedom from rejection (FFR) rates for sensitized and control recipients, 53.3% and 66.7% respectively, p = 0.367 (Figure 2). Thereby, virtual crossmatching appeared to facilitate a reduction in immunologic risk in the sensitized recipients to that of the control recipients.

Table 4. Rejection data
 Control (n = 27)Sensitized (n = 15)p-Value
  1. 1Mean ± SD.

  2. 2Several patients (n = 4) required more than one therapy per rejection episode.

Number of biopsies122 ± 1025 ± 110.405
Acute rejection (AR)—no.
 Total number of episodes11 (40.7)10 (66.7)0.243
 Grade 13 (11.1)2 (13.3)0.400
 Grade 23 (11.1)4 (26.7)1.000
 Grade 35 (18.5)4 (26.7)0.358
Treatment per rejection episode—no.2
 Steroids only6 (22.2)5 (33.3)0.433
 Thymoglobulin4 (14.8)6 (40)0.155
 Campath1 (3.7)2 (13.3)0.287
 Infliximab01 (6.7)0.357
Time to 1st rejection (days)1167 ± 218107 ± 790.435
Duration of rejection (days)118 ± 14 36 ± 270.139
Chronic rejection—no. (%)01 (6.7)0.357
Explant—no. (%)1 (3.7)3 (20)0.122
Figure 2.

One-year freedom from rejection in sensitized compared to nonsensitized recipients. Kaplan–Meier curve with p-value by log-rank test.

Patient and graft outcomes

Patient and graft outcomes were compared between sensitized and control groups. TPN days, ventilator days, length of index admission, number of readmissions, overall perioperative infections, incidence of PTLD and 1-year cause of death were not significantly different between the two groups (Table 5). The incidence of spontaneous perforation was significantly higher in the sensitized recipients, which is thought to be immune-mediated, but is poorly understood. There was no statistical significant difference in 1-year patient survival in sensitized compared to control recipients (73.3% and 88.9% respectively, p = 0.197) and 1-year graft survival (66.7% and 85.2% respectively, p = 0.167) (Figures 3 and 4). Overall, patient and allograft outcomes were comparable between the sensitized and control groups.

Table 5. Complications
 Control (n = 27)Sensitized (n = 15)p-Value
  1. 1Mean ± SD.

  2. 290-day perioperative infection, many patients had more than one infectious episode.

TPN days118.8 ± 20.517.3 ± 9.9 0.809
Ventilator days12.2 ± 1.72.1 ± 2.10.799
Length of stay (days)136.2 ± 32.151.5 ± 69.10.428
Readmissions14.2 ± 3.22.9 ± 2.40.198
Perioperative infection—no. (%)2
 Peritonitis4 (14.8)5 (33.3)0.242
 Intraabdominal abscess7 (25.9)4 (26.7)1.000
 Bacteremia6 (22.2)7 (46.7)0.163
 Fungemia2 (7.4)3 (20)0.608
 Pneumonia5 (18.5)2 (13.3)1.000
 Wound infection4 (14.8)1 (6.7)0.639
Spontaneous perforation—no. (%)1 (3.7)4 (26.7) 0.047
PTLD—no. (%)4 (14.8)1 (6.7)0.639
1-yr cause of death—no. (%)
Rejection01 (6.7)0.357
Sepsis1 (3.7)2 (13.3)0.287
PTLD1 (3.7)1 (6.7)1.000
GVHD00
Thrombosis1 (3.7)01.000
Figure 3.

One-year graft survival in sensitized compared to nonsensitized recipients. Kaplan–Meier curve with p-value by log-rank test.

Figure 4.

One-year patient survival in sensitized compared to nonsensitized recipients. Kaplan–Meier curve with p-value by log-rank test.

Discussion

In this study, we describe the first experience with a VXM organ allocation strategy to reduce immunologic risk in disadvantaged sensitized recipients of isolated intestine transplants. With the use of VXM, the majority of sensitized recipients were transplanted with a negative or weakly positive flow-cytometry XM (FC-XM) and with zero or only low-titer (≤1:16) DSA. Organ availability was also increased in sensitized patients, with comparable wait-list time and cold ischemia time to that of the control patients. Importantly, patient and graft outcomes were not compromised, as there was no statistical difference in 1-year acute rejection, graft and patient survival in sensitized compared to control recipients. The VXM was successful in enabling transplantation of the sensitized isolated intestine transplant patient.

Intestine transplantation has advanced rapidly in the last decade and is currently the preferred therapy for patients with intestine failure and total parenteral nutrition (TPN)-related complications (20). However, the immunologic barrier to successful intestine transplantation remains formidable. Despite advances in immunosuppression protocols and aggressive endoscopic surveillance, acute rejection is common and often associated with allograft loss or patient death. Currently, the incidence of acute rejection for the isolated intestine graft is as high as 50% in most large series and remains a major clinical challenge in intestinal transplantation (15,21–23).

Sensitizing events are common in patients referred for intestinal transplant, as these patients universally undergo multiple operations with concomitant blood transfusions, suffer from recurrent line infections and in some cases require retransplantation. High panel reactive antibody (PRA) levels are found in 18–30% of patients on the waiting list for intestinal transplantation, a high rate compared with other solid organ transplant recipients (3–5). Indeed, in our series the incidence of sensitization was significant at 36%. Sensitization increases the risk of rejection in an already highly immunogenic transplant, especially with a positive CDC-XM (1,2,24). Altogether, the sensitized isolated intestine recipient represents the highest immunologic risk transplant patient.

The liver is relatively resistant to the effects of preformed antibody-mediated humoral rejection (25), and the inclusion of a liver graft with the intestine may be immunologically protective in the sensitized recipient (21,26,27). However, the requirement for a liver inclusion graft is decreasing (2011 Intestine Registry Report, http://www.intestinetransplant.org), likely due to the improved management of intestine-failure associated liver disease (28,29). Concurrently, intestine transplant outcomes continue to improve and 1-year patient survival rates now exceed 90% at experienced high-volume intestinal transplant centers (15,23,30). With the growing success of intestinal transplantation, more patients will undergo this therapy, and the isolated intestine transplant will likely become the predominant procedure. As such, management of the sensitized isolated transplant recipient will become increasingly relevant to intestinal transplantation.

As in other solid organ transplants, sensitization increases the risk of a positive crossmatch, due to lymphocytotoxic antibodies specific to the intestine recipient (1,2,24). A positive CDC-XM has been shown to significantly increase the risk of acute rejection and graft loss following intestinal transplantation (26). As first described by Wu et al., a syndrome of mucosal congestion, cyanotic discoloration and focal hemorrhage within the allograft was observed in five patients undergoing CDC crossmatch-positive isolated intestine transplants (5). These patients required aggressive rescue immunosuppression with OKT3 and high-dose steroids. In another early study, a positive CDC-XM was associated with increased frequency and severity of acute intestinal rejection, particularly in isolated intestine transplant recipients (3). Based on these early studies, CDC-XM positive isolated intestine transplantation has generally been avoided due to the prohibitive immunologic risk. However, with limited organ availability and the relatively short acceptable cold ischemia time, obtaining a crossmatch-negative isolated intestine transplant in the sensitized recipient is extremely challenging. The introduction of the VXM has provided an opportunity to transplant this difficult population.

The available technology for the detection and therapeutic manipulation of lymphocytotoxic antibodies has greatly advanced since the first descriptions of humoral rejection in kidney transplantation (31,32). Characterization of lymphocytotoxic antibodies has evolved from the basic CDC-XM to the more sensitive FC-XM, and now to the highly specific identification and quantification of single HLA antibody with the use of solid phase assays. Solid-phase assays have made it possible to perform a VXM, in which DSA can be determined ‘virtually’ by comparison of known HLA specific antibodies to donor HLA-type (33). The VXM can be performed prospectively and with at least similar sensitivity to the FC-XM (9,34). In most studies, the correlation of the VXM with the FC-XM was greater than 85% (6,8,35,36). In addition, multiple recent kidney and thoracic transplant studies have reported that a negative VXM was associated with a very low risk for early rejection and good allograft survival even in sensitized recipients (6,9–13).

The VXM has been shown to accelerate organ allocation in the disadvantaged sensitized patient (6–8). Heart transplant is most similar to intestine transplant in that the allowable cold ischemia time is limited and significant travel is often required for high quality organs. In a recent study by Yanagida et al., a VXM approach was used to transplant sensitized heart recipients (7). The authors demonstrated reduced waiting times to transplant and the ability to travel outside of their region to obtain optimal organs without compromising CIT. In the current study, VXM was successfully employed for organ allocation in sensitized intestine transplant recipients. Sensitized recipients were transplanted with comparable wait-list time and CIT to nonsensitized recipients. Our results demonstrate that the VXM essentially eliminates the risk of forfeiting an otherwise acceptable intestine donor due to a positive CDC-XM and normalizes the wait-time in sensitized patients to that of nonsensitized patients. Also, with the prospective VXM, the recipient does not wait for the conventional crossmatch which reduces CIT, despite considerable distance traveled for the donor intestine allograft.

The VXM can also be used to reduce the immunologic risk in sensitized transplant recipients. Several recent studies have reported that a negative VXM based on single-antigen assays is associated with a very low risk for early rejection and good allograft survival in sensitized solid organ transplant recipients (6,9–12,37). For example, in the Yanagida study, sensitized heart recipients were transplanted without increasing the subsequent occurrence of cellular rejection, antibody mediated rejection and mortality (7). Our results also demonstrate that VXM can be used to transplant sensitized isolated intestine recipients without significantly compromising graft and patient outcomes. In our study, there was no significant difference in 1-year freedom from rejection in sensitized and control recipients. The VXM minimizes the incidence of a positive crossmatch and allows sensitized patients to be transplanted with similar immunologic risk compared to control patients.

Ideally, no patient would be transplanted with positive DSA. However, given the severely limited supply of suitable intestinal allografts, low levels of DSA (≤1:16) was considered acceptable in these sensitized and difficult-to-transplant patients. This threshold has been studied in living donor kidney transplantation and generally represents a DSA quantity with a low-risk of a positive crossmatch and that is amenable to therapeutic manipulation with IVIG and/or plasmapheresis (38–40). Although the role of DSA in intestine transplantation is poorly characterized, DSA has recently been identified as a risk factor for acute rejection in intestine transplantation. In this study, Tsai et al. found that preformed DSA was associated with more frequent and severe acute rejection episodes, graft loss and mortality in a cohort of 13 recipients of intestine transplants (41). As DSA emerges as a potential risk factor for rejection in intestine transplantation, we are currently developing a protocol for routine posttransplant DSA monitoring in sensitized recipients.

One-year patient and graft survival were decreased in the sensitized compared to control groups. Although not statistically significant, this may be clinically significant and deserves comment. It should be emphasized that prior to the application of the VXM, it was extremely difficult to obtain a crossmatch negative transplant for the sensitized isolated intestine patient. Consequently, many of these patients simply did not get transplanted. Patients who have failed TPN therapy have a high mortality rate, as is evident by the mortality rate of the intestine transplant waiting list (42). Indeed, patients awaiting intestinal transplants in the United States have the highest mortality of any group on the solid organ waiting list maintained by the UNOS registry (available at http://www.unos.org), up to 25% for adult recipients and 60% for pediatric recipients (43). In summary, although 1-year patient and graft survival were somewhat decreased in our sensitized recipients, without VXM these patients may not have had the opportunity to be transplanted, and likely would have died of lethal TPN-related complications.

This study has several important limitations. This was a retrospective study of a VXM practice that is still in evolution at our institution. In particular, we are increasingly using the prognostic capability of the VXM to individualize immunosuppression decisions in the sensitized recipient. For example the use of thymoglobulin and IVIG is now generally reserved only for sensitized recipients with a positive FC-XM or DSA. This is important, as our group and others have identified thymoglobulin as a significant risk factor for the development of PTLD (44). Another limitation of this study was the lack of consistent DSA monitoring posttransplant in our sensitized recipients. Posttransplant DSA was only sporadically obtained for-cause rather than on protocol. As mentioned, we are currently developing a DSA monitoring protocol for all sensitized recipients. We also did not routinely obtain C4d staining of biopsies in sensitized recipients as the clinical significance of C4d is uncertain in intestine allografts (45,46). Although there is no consensus definition for antibody-mediated rejection in intestine transplantation, without robust posttransplant DSA and C4d data, it is difficult to acertian the incidence of antibody-mediated events in our study population. Humoral rejection may be a clinical entity distinct from cellular rejection in intestinal transplantation, but requires defined criteria before it can be rigorously studied. It should also be noted that the VXM is not infallible, and there can be false-positive or false-negative results (47–49). For example, the VXM may be false negative due to incomplete HLA typing of the donor, as HLA typing for deceased donors often does not include HLA-DP (50,51). Finally, although we demonstrate comparable short-term outcomes in sensitized compared to control recipients, longer follow-up is necessary to determine the late immunologic sequela of sensitization in isolated intestinal transplant.

In conclusion, we describe the successful application of VXM to transplant sensitized isolated intestine recipients. The VXM accelerated organ allocation in the disadvantaged sensitized recipient, with comparable short-term graft and patient outcomes to nonsensitized recipients. As intestine transplantation outcomes continue to improve and the isolated intestine allograft is increasingly utilized, immunologic risk reduction in the sensitized patient will become increasingly relevant. In this study, we demonstrate a VXM strategy to successfully transplant this immunologically high-risk population.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Ancillary