Blood and Marrow Transplantation Beyond Barriers

Introduction

Allogeneic Blood and Marrow Transplantation (BMT), which employs the use of hematopoeitic progenitor cells (HPC) from a donor to replace the diseased or defective hematopoeitic as well as the immune system has evolved radically over the last three decades. The indications for Allo-BMT are no longer limited to end-stage leukemias. This is deemed curative and often the only option for many non-malignant conditions as well. These advances have been possible through better understanding and management of post-transplant complications, both infectious and non-infectious. The key to this success is perhaps related to the optimum selection of donors for Allo-BMT. Human Leukocyte Antigen (HLA) system is pivotal to the outcome of Allo-BMT unlike solid organ transplantation. HLA-matched family member was and still is considered the donor of choice for Allo -BMT for all indications; a paradigm which remains unchanged despite tremendous advancements in this field. Based on mendelian laws of inheritance, the chances of siblings being matched to each other at class I and II HLA locus are only 20-30%. Thus 70-80% of patients eligible for an Allo-BMT would be unlikely to find a donor within the family. Thus the large majority of patients does not find a matched family donor and are candidates for an alternative donor BMT.

Unrelated Donors as a Source of Cells for Allo-BMT

The options for alternative donor were limited by the initial studies demonstrating the primacy of HLA matching in dictating the outcome of BMT.  The chances of finding a HLA match within the extended family is remote and even more so from an unrelated donor. The National Bone Marrow Donor Registry was federally funded in 1986, and in 1987 the first donor match was made. In 1988, the name was changed to the National Marrow Donor Registry (NMDP). The NMDP now includes a network of donor registries in 30 countries. Its database contains more than 5.5 million donors and facilitates an average of 200 transplants each month. Several such registries are operational in all developed countries accounting for over 11 million donors worldwide. At the inception, UDBMT was an extremely high risk procedure with a transplant related mortality (TRM) of nearly 50 % which is now reduced to less than 30%.

Unrelated Cord Blood Transplantation (UCBT)

A decade after the first UDBMT, Gluckman and Broxmeyer carried out the first successful related cord blood transplantation (CBT) in a patient with Fanconi's Anemia. This was a result of several years of research establishing the survival and repopulating potential of cord blood both fresh and frozen. It was established subsequently that compared to bone marrow, one log less mononuclear cells are sufficient for engraftment in CBT, but this is invariably delayed.

Whilst cord blood provided a unique source of HPC, engraftment was uniformly delayed for neutrophils and more so for platelets. Since the first CBT, more than 20,000 CBTs have been reported worldwide and more than 400,000 cord blood units have been stored in more than 100 cord blood banks. The main practical advantages of using cord blood as an  alternative source of stem cells are the relative ease of  procurement, the absence of risks for mothers and donors,  the reduced likelihood of transmitting infections, particularly  CMV and the ability to store fully tested and HLA-typed  transplants in the frozen state, available for immediate use.

The donor who is always there- the Haploidentical Family Donor

Based on Mendelian laws of inheritance, HLA alleles are inherited as haplotypes from each parent. Thus, parents are matched at at least one haplotype with the patient. In addition, 90% of the siblings are expected to be matched at one haplotypes. Initial attempts at transplantation with haploidentical donors within the family was met with universal failure as a result of very high incidence of graft rejection or severe GVHD.  With the developments in UDBMT and UCBT, this donor source was ignored completely in the mainstream of AlloBMT.

A group of researchers led by Martelli and Reisner, proposed a modality of Haploidentical BMT wherein infusion of very high dose of CD34 + cells depleted of T cells would be infused following intensive myeloablative as well as immunoablative conditioning. It would take several years to optimise the procedure and in its final form, CD34 + cells would be purified using immunomagnetic methods for positive selection and the final product would contain less than 2 x 10 4 /kg T cells and more than 10 x 10 6/ kg CD34 cells. Almost all patients achieved engraftment with virtually no GVHD.

NK Cell Alloreactivity

On further exploration, the same group uncovered an unique phenomenon related to superior survival of patients with AML. Pioneering work by Ruggeri et al demonstrated that, in the setting of  haploidentical transplantation, NK-cell alloreactivity is key to the antileukemia effect. Those patients who were mismatched at Killer Ig Receptor (KIR) genes of the NK cells with the donor had a significantly superior survival without any increase in GVHD. This was the first time GVHD and GVL could be dissected at a clinical level.

Standard GVHD Prophylaxis

The researchers in China adopted an intensive conditioning schedule with aggressive GVHD prophylaxis along with G-CSF primed marrow or peripheral blood. The results from Peking University were particularly impressive. 100% engraftment was achieved in 250 patients with acute leukemia. In addition, Haploidentical donor graft conferred a survival advantage over matched donors in advanced leukemia due to a superior GVL effect.

Post transplant High Dose Cyclophosphamide

Cyclophosphamide was one of the first drugs used in conditioning for BMT. It was also  tried as GVHD prophylaxis in low doses due to its immunosuppressive properties. However this was not found to be effective at the low weekly doses employed in these studies. Although the preclinical data confirmed the efficacy of cyclophosphamide in inducing transplant tolerance at a much higher dose, the clinical studies employed much lower doses due to the concern over its deleterious effect on progenitor cells. The researchers at the Johns Hopkins and others discovered the unique property of primitive HPCs of being resistant to the cytotoxicity of cyclophosphamide by dint of very high expression of aldehyde dehydrogenase which metabolises the drug.

The Indian Scenario

Compared to over 10,000 BMT performed in the USA per year, less than 1000 are performed in India as of 2010 with only 5% being from alternative donors compared to 50% in the USA. There is no public cord blood bank in India and two struggling Unrelated Donor registries. The chance of obtaining a 10/10 HLA match for an Indian donor from European or American registries is only about 10%. Moreover the cost of obtaining an UD graft for an Indian patient is about $25-30,000. The same is the case with UCBT. Thus most patients eligible for an allogeneic BMT do not receive a transplant in India. One must learn from the outcome of African-Americans undergoing UDBMT. The realistic possibility of obtaining a 10/10 matched unrelated donor from a Caucasian dominated registry is about 6%. Repeated studies have confirmed poorer outcome after UDBMT in this group which is attributable to less engraftment and more GVHD related to less than full match and transplants across ethnic groups. Similar inferior outcomes have been reported following mismatched UCBT in African-Americans.

Problems Thwarting Development of Unrelated Donor Registry (UDR)  in India

The absence of a congruous policy and access to healthcare for all it's citizens plague the healthcare system in India. Development of UDR needs education, awareness and motivation amongst the people along with a sense of altruism. Lack of a national policy regarding cancer and thalassemia does not allow further headway regarding more intensive therapy regarding these diseases, such as BMT. Where voluntary blood donation is linked to guaranteed availability of blood products to the donor, volunteer bone marrow donor registry is likely to be a more distant reality. Several attempts have been made by various organisations, but poor recruitment and a faster attrition of donors have hampered the development of a viable registry. In addition, the funding required for infrastructural support and sustenance is a daunting ask for charitable organization.

On the other hand, Public Cord Blood Bank requires a lot more capital investment although there is no risk of attrition. Conservatively, a UDR requires one million donors and a PCBB should have an inventory of at least 20,00 units if more than 50% of patients are to find a suitable alternative donors. Either of these requires active BMT programs across the country to sustain such alternate donor banks with over 500 UDBMT and UCBT performed per year. Until that becomes a reality or even without it, the need for Indian patients requiring alternative donor BMT is to develop Haploidentical BMT Programs.

Haploidentical BMT in the Indian context

Although a few of the centres in India are carrying out UDBMT and UCBT, none have developed a Haploidentical BMT program. In view of the above consideration, we developed a Haploidentical BMT program first time in India. In the first phase, we selected only patients with refractory AML. The donor selection was based on the following algorithm: 1) Mother if NK alloreactive donor is not available 2) NIMA mismatched donor if mother is not available as a donor.  All patients were transplanted with over 30% marrow blasts and had failed several lines of treatment. The conditioning protocol employed high dose cytosine and an anthracycline for 3 days followed by a regimen similar to Johns Hopkins except low dose TBI being replaced by melphalan. High dose cyclophosphamide was employed on days 3 and 4 after infusion of the graft which was mobilized peripheral blood HPC rather than bone marrow.

All 3 patients engrafted neutrophils and platelets at a median of 14 and 16 days respectively. Donor engraftment above 95% was achieved in all at 30 days. None developed acute GVHD de novo. At a median follow-up of 180 days, all had relapsed. Our study suggested that Haploidentical BMT is feasible in patients with advanced leukemia. The prompt engraftment and lack of GVHD warranted Haploidentical BMT to be employed early in the course of the disease and with more intensive conditioning. The subsequent 8 patients were transplanted with more intensive conditioning with 50% disease free survival at one year.

We extended this program further to patients with Severe Aplastic Anemia (SAA) and thalassemia. 8 patients with Aplastic anemia and one patient with Thalassemia32 have undergone Haploidentical BMT with 60% survival at one year. Our research on the NK cell alloreactivity in Haploidentical BMT using post-transplant cyclophosphamide has brought to light several key issues related to SAA and its outcome for the first time.

More importantly, in our Haploidentical program with over 25 Haploidentical BMT performed, the cost of such BMT was comparable to that of matched family donors. In our attempt to expand the program to the next level, the newer technologies of selective T cell depletion are going to be introduced, which would be more expensive but expected to be more successful at the same time with lesser complications and virtual elimination of long term immunosuppression and antibiotic usage. UDBMT from western registries is unlikely to be a long term solution due to both the cost involved and a low yield of full HLA matched donors. The same is true for UCBT.  Mismatched UDBMT should be discouraged from ethnically disparate registries if we are to learn from the outcome in African-Americans. Similar to other Asian countries such as China and Korea, Haploidentical BMT programs need to be developed in India to cater to the majority of patients requiring a BMT. The challenge in India is not just to develop a Haploidentical BMT program for malignant diseases, but to do the same for nonmalignant diseases which account for half of the transplant indications currently.