Highly Successful and Low-Cost Desensitization Regime for Sensitized Living Donor Renal Transplant Recipients

Abstract

10–30% of dialysis population awaiting renal transplantation is sensitized. Present desensitization protocols use intravenous immune globulins, rituximab, and plasmapheresis in various combinations; however, these regimens are unaffordable by many in developing countries. We tried desensitization with mycophenolate mofetil and plasmapheresis. Methods. Patients with high PRA titre (≥50%) or positive crossmatch (>10%) were treated with MMF for a month before proposed transplant and were given five sittings of plasmapheresis. Results. 11 of 12 patients had normalization of PRA/crossmatch with this regimen and were successfully transplanted. One patient lost the graft due to graft vein thrombosis, and two patients died within three months after transplant due to septicemia and pulmonary embolism, respectively, with a functioning graft. No patient, including the two who died, developed clinical rejection over a mean follow-up of 10 months (range 1–16 months). Mean serum creatinine at last follow up was 1.1 mg/dL (range 0.9–1.3 mg/dL). Conclusions. Though the number of patients studied is small, we feel that highly sensitized patients awaiting living donor renal transplant should be tried on this simple and cost-effective regime before transplant. The more aggressive and expensive approaches incorporating IVIg and rituximab should be used only if this relatively low-cost regime is unsuccessful.

Keywords:

• highly sensitized
• PRA
• positive crossmatch
• desensitization

INTRODUCTION

Sensitized patients have a variety of antibodies that affect graft survival (e.g., antibodies against HLA class I and II antigens, antibodies against B cells, anti-endothelial antibodies, or antibodies against MHC class I gene).[1–3] Of these, the antibodies against HLA class I antigens are probably the most important and are most frequently monitored. Panel reactive antibodies (PRA) and cross match (XM) are the tests used for detection of HLA class I antibodies. By convention, patients having PRA levels of >50% are called highly sensitized. Previous transplant, pregnancy, and blood transfusion are the usual causes of sensitization. 20–30% of patients awaiting cadaveric transplant and about 10% of patients awaiting living donor transplant are sensitized.[4] Many of them have PRA levels >80% and have to wait for prolonged periods for successful transplantation. Even after transplant, some of them develop hyperacute rejection or antibody mediated rejection (AMR) with significant graft loss.[4,5] Many centers have developed their own desensitization protocols based on combination of high-dose or low-dose intravenous immunoglobulin (IVIg), plasmapheresis, and rituximab.[5–7] However, such regimens are expensive and unaffordable by most patients from developing countries who can barely afford the cost of transplant in absence of state-sponsored programs. In addition, such protocols can predispose patients to infections, polyoma virus infection, and neoplasia.[8] Mycophenolate mofetil (MMF) inhibits T and B cells by inhibiting rate-limiting enzyme inosine-5 monophosphate dehydrogenase.[9] We tested this property of MMF to inhibit antibody production in highly sensitized patients.

METHODS

The subjects were comprised of sensitized patients awaiting living donor transplantation with PRA >50 % or positive XM done on at least two occasions a month apart. PRA was done by ELISA technique using commercially available kit (GTi Diagnostics, Quik-ID Class I), class I antigens being the target antigens. All antigens present in the donors were well-represented. XM was done with complement dependent cyto-toxicity (CDC) method, using standard NIH technique with class I antigens as the target antigens. Patients were started on MMF 1 gm twice a day a month prior to proposed transplant, after informed consent was taken. Three weeks after MMF therapy, patients were given five sittings of plasmapheresis over 7–10 days. During the five sessions of plasmapheresis, we used a total of two plasma filters, a unit of albumin, and 6–8 units of fresh frozen plasma at each sitting. Meanwhile, MMF was continued uninterrupted.

The PRA titers were repeated after three weeks of MMF therapy in the first three patients enrolled in the study, and though the titers reduced from 100%, 65%, and 65% to 85%, 55%, and 60%, respectively, these levels were still unacceptable. Hence, these patients were subjected to five sessions of plasmapheresis, and in all of the subsequent patients, PRA levels were repeated only after completion of the four weeks therapy, inclusive of the five sessions of plasmapheresis, and four weeks of MMF therapy. Patients who became cross match-negative or had acceptable levels of PRA were transplanted. Induction was done with anti-thymocyte globulin (ATG) 10 mg/kg/day for 10 days or with IL-2 blockers. The immunosuppressive protocol consisted of tacrolimus (blood levels 10–12 ng/mL), MMF 1 gm BD, and prednisolone 0.5 mg/kg/day. All patients were discharged after a fortnight of transplant and subsequent follow-up was as per our center's protocol (i.e., thrice a week during initial 15 days, twice a week over next 15 days, weekly for next two months, fortnightly from 4–6 months, and monthly thereafter till 12 months). Tacrolimus levels were maintained between 5–8 ng/mL after three months. Prednisolone taper was started after a month, and at the end of three months, patients were receiving 5–10 mg/day, a dose that was maintained thereafter. Any patient who developed unexplained graft dysfunction underwent biopsy. As per our center practice, a protocol biopsy was done at 3 and 6 months post-transplant period in all patients who did not develop clinical rejection. Rejection if any was treated with 0.5 gm of methyl prednisolone per day for 3–5 days.

RESULTS

Twelve of 102 prospective renal transplant recipients were found to be sensitized. The relevant clinical details and results of PRA and XM of these cases are shown in Table 1. In one of these patients, PRA could not be done; however, his XM was positive. There were 9 males and 3 females with mean age of 40 years (range 23–58 years). Mean PRA of the 11 patients was 74.31 + 20.15 (range 40–100). Five patients had positive XM with mean value of 56 % (range 20–100). After a month of MMF therapy, 10/11 (91%) had PRA of <10% (p < 0.001, paired t test); however, one patient had no response to therapy, and his PRA remained almost same (97.5% from baseline of 100%). All patients tolerated MMF well without any significant side effects or requirement for reduction in dose or withdrawal of the drug. All five patients with a positive XM became XM negative at the end of the protocol. The mean waiting time to transplantation was five months (1–18 months), and 11/12 patients were transplanted from living donors, the donors being father in 1, sister in 2, son in 1, and spouse in the remaining 7 (wife in 6 and husband in 1). Four patients received IL-2 blockers while the 7 with spousal donors received ATG as induction therapy. One patient developed graft vein thrombosis and lost his graft, while the other 10 achieved normal graft function. During the follow-up period, no patient developed clinical rejection. However, on protocol biopsy at six months, two patients were found to have grade 1b rejection and received steroids though their creatinine was normal. Two of these 10 patients died two and three months after transplantation due to septicemia and pulmonary embolism, respectively, but with a functioning graft. No other patient had any major infectious complication during the follow-up period. The mean creatinine in the remaining eight patients after a mean follow-up period of 12 months (range 1–17 months) was 1.1 mg/dL (range 0.9–1.3 mg/dL). The HLA typing of the recipients and donors is shown in Table 2. Seven of the 11 cases had spousal donors, and in these, only HLA class I antigens were done. Of these 7 cases, 4 had 0/4 match, 2 had 1/4 match, and 1 had 3/4 match. In the remaining four patients with related donors, both HLA class I and II antigens were done, and all were haploidentical with 3/6 HLA match.

Table 1 Demographic, pre- and post-treatment PRA and XM, and post-transplant follow-up details of the study group

S Number	Age	Sex	Donor	HLA match	Pre-treatment PRA in %	Pre-treatment XM in %	Post-treatment PRA in %	Post-treatment XM in %	Date oftransplant	Outcome	Rejection	Last creatinine	Follow-up (in months)
1	30	M	Sister	3/6	100	<10	2.5	<10	13.12.06	Well	None	1.3	17
2	37	M	Wife	3/4	65	<10	0	<10	3.01.07	Well	None	1.2	16
3	23	M	Father	3/6	65	<10	10	<10	21.02.07	Well	None	1.1	15
4	58	M	Wife	1/4	ND	40	7.5	<10	07.03.07	Died after three months of pulmonary embolism	None	1.6	3
5	29	F	Sister	3/6	85	100	0	<10	11.04.07	Well	None	1.1	13
6	40	M	Wife	0/4	100	<10	97.5	<10	On dialysis	—	—	—
7	50	M	Wife	0/4	40	20	0	<10	20.04.07	Graft vein thrombosis	—	Graft loss	0
8	42	F	Son	3/6	72.5	100	2.5	<10	25.04.07	Well	On PB	1.2	13
9	39	M	Wife	1/4	52.5	<10	0	<10	6.06.07	Well	On PB	0.9	11
10	38	F	Husband	0/4	100	<10	0	<10	29.06.07	Well	None	1.0	11
11	49	M	Wife	0/4	75	<10	7.5	<10	18.07.07	Died after two months of septicemia	None	2.3	2
12	45	M	Wife	0/4	62.5	20	0	<10	7.05.08	Well	None	1.2	1

*Not transplanted due to persistently high PRA.

Abbreviations: PB = protocol biopsy, ND = not done.

Table 2 HLA profile of recipients and donors

S Number	Donor relation	Recipient						Donor						HLA match
		A	A	B	B	DR	DR	A	A	B	B	DR	DR
1	Sister	3	11	13	60	5	13	33	11	35	60	7	13	3/6
2	Wife	3	11	51	61	ND	ND	3	11	51	49	ND	ND	3/4
3	Father	3	31	51	8	14	15	3	3	51	51	3	14	3/6
4	Wife	31	33	27	57	ND	ND	1	2	27	81	ND	ND	1/4
5	Sister	11	26	52	13	15	11	11	28	52	57	15	3	3/6
6	Wife	34	33	51	51	ND	ND	3	26	27	61	ND	ND	0/4
7	Wife	23	23	44	35	ND	ND	1	11	57	57	ND	ND	0/4
8	Son	3	3	51	63	12	15	2	3	60	63	12	13	3/6
9	Wife	26	32	8	51	ND	ND	1	66	8	37	ND	ND	1/4
10	Husband	2	28	51	8	ND	ND	9	11	35	35	ND	ND	0/4
11	Wife	1	28	44	57	ND	ND	1	11	51	62	ND	ND	0/4
12	Wife	11	24	35	61	ND	ND	3	3	27	27	ND	ND	0/4

*Not transplanted due to persistently high PRA.

Abbreviation: ND = not done.

DISCUSSION

Sensitized patients can have a variety of antibodies (e.g., HLA class I and II antibodies, anti-endothelial antibodies, and antibodies against MHC class I gene). Mizutani et al.[1] reported association of all these antibodies independently with slow graft failure. However, the most important of these are IgG anti HLA class I antibodies. About 20–30% of prospective cadaver transplant recipients and approximately 10% of prospective living donor transplant recipients are sensitized and have to wait for years before transplant.[2] Cross match and PRA are the usual tests performed for these antibodies. Detection of these antibodies can be done by various methods, such as complement dependent cytotoxicity (CDC), CDC-with anti-human globulin (CDC-AHG), ELISA, or flow cytometry cross match (FCXM). Though not universally available, FCXM is the most sensitive of these methods.[10] Karapinski et al. observed that though negative by CDC-AHG, 18 of the 143 patients (13%) were found to have positive FCXM, and 33% of them had early graft loss due to acute humoral rejection.[11] Hence, few sensitized cases can be missed if XM is done by CDC or ELISA methods. In a prospective study over four years, Terasaki et al. reported poorer graft outcome in patients with HLA or MICA antibodies.[2] Similarly, Barama et al. reported inferior graft outcome at 1, 3, and 5 years in sensitized patients.[3] Survival was poorer as the sensitization level rose, and 20% of grafts were lost in the initial six months in the group having PRA between 50–100%. Zhou and Cecka have also reported 7% inferior graft survival in sensitized patients.[4] The impact of HLA class II antibodies on graft survival is debatable, though some researchers have shown that the presence of donor-specific HLA class II antibodies is detrimental to graft survival.[12,13]

NIH IG 02 trial showed the efficacy of IVIg in desensitization.[14] In this trial, IVIg 2gm/kg was given for four consecutive months to highly sensitized patients (PRA >50%). IVIG significantly reduced the antibody levels (p = 0.033) and time to transplantation (p = 0.049). IVIg has multiple functions, like neutralization of the antibodies, anti-cytokine properties, inhibition of proliferation of T & B cells, and inhibition of antibody formation.[15] The basic principle of all protocols used for desensitization is to inhibit the formation of antibodies and neutralization or removal of the existing antibodies. Half of these antibodies are present in intravascular compartment, and an equal amount is in extravascular compartment. Plasmapheresis removes more than half of intravascular compartment antibodies, but in the next 24 hours, antibodies shift from extra- to intravascular compartment.[16] Therefore, the fundamental issue should be to stop or inhibit formation of new antibodies. This has been tried with cyclophosphamide and rituximab. Many centers have devised their own desensitization protocol with fairly good results.[5–7] All of these protocols are based on the use of IVIg/plasmapheresis/rituximab and immunosuppressives. Glotz et al. could transplant 13/15 highly sensitized patients (PRA >50% or positive cross match) following three monthly doses of IVIg.[15] Two of their patients lost the graft due to AHR, while the others maintained normal graft function.

We completed XM by serological method, and it is possible that if ELISA/FCXM was done, more patients could have been XM-positive. We went with the analogy that we stop the antibody production by inhibiting B cells with MMF and thereafter remove the existing antibodies with plasmapheresis. To the best of our knowledge, no center has tried an MMF-based protocol. Though the number of patients in our study is small, it shows good results with a cost-effective protocol. In India, very few patients can afford transplantation, and the desensitization protocol based on IVIg and rituximab will be unaffordable by most patients. The cost of our regimen is less than $2500 per patient, which is affordable. We recommend that in living donor transplant program, MMF and plasmapheresis should be tried in highly sensitized patients. The more aggressive protocols should be used only if this relatively low cost regime is unsuccessful.

ACKNOWLEDGMENTS

The authors report no conflict of interest, and are alone responsible for the writing of this article.