Benefit of STR-based chimerism analysis to identify TA-GVHD as a cause of death: Utility of various biological specimens

Introduction

Transfusion-associated graft-versus-host disease (TA-GVHD) is a fatal though rare complication of blood transfusion.^1,2 In 1955, Shimoda ³ observed postoperative erythroderma (POE) and reported TA-GVHD for the first time wherein 12 patients had developed skin rash and fever which began 6–13 days after surgery. Six of these patients died; the other six survived after treatment with antibiotics and steroids. TA-GVHD is characterised by fever, skin rashes, diarrhoea and pancytopenia. ⁴ The diagnosis of TA-GVHD is often delayed because of a lack of awareness and non-specific clinical features. Although it is commonly associated with immune-compromised hosts, it has also been reported on immune-competent patients, particularly from homogenous genetic populations and patients receiving transfusions from first-degree relatives. TA-GVHD is caused by the attack of the viable donor lymphocytes on the recipient. This results in engraftment of the transfused cells which subsequently leads to rejection of host cells due to their immunologic differences. Clinically, TA-GVHD has some features of bone-marrow/stem cell-transplantation-associated GVHD. Bone-marrow-transplant-related GVHD occurs weeks to months after bone marrow transplant (BMT) and presents with an erythematous, maculopapular rash, elevated liver enzymes often associated with hepatomegaly and jaundice plus gastrointestinal symptoms including nausea, vomiting and diarrhoea. TA-GVHD presents with similar symptoms, but their onset is usually earlier, within 7–10 days of transfusion.^5,6 However, the onset may be insidious, and the exact correlation with transfusion is often missed clinically. Hence, most diagnoses are made post-mortem. The mortality rate of TA-GVHD is very high (80–90%) compared to BMT-associated GVHD mortality (20–25%).

Case report

An 11-month-old male child with congenital cytomegalovirus (CMV) infection (diagnosed on the basis of positive blood and urine real-time polymerase chain reaction [PCR]) was admitted to the All India Institute of Medical Sciences (AIIMS), New Delhi, for treatment with gancylovir. He continued to have bleeding from various sites, progressive pallor and skin rashes. He had received several blood transfusions prior to hospitalisation.

In terms of perinatal history, the baby was born preterm at 35 weeks of gestation, with a very low birth weight (1110 g), delivered by lower-segment Caesarean section (LSCS) and was admitted to the neonatal intensive care unit (ICU) for 21 days at AIIMS.

Physical examination of patient showed his weight and height to be below the third percentile and also revealed pallor and hepatosplenomegaly. During the hospital stay, the patient was noted to have dryness and scaly skin lesions.

On investigation, the patient had severe anaemia (haemoglobin [Hb] 6.1 g/dL) and thrombocytopenia (platelet count 30 × 10⁹/L). The patient received multiple peripheral red blood cells (PRBC) and platelet transfusions during his hospital stay. Differential diagnosis of Langerhans cell histiocytosis (LCH), Evan’s syndrome and unresponsive CMV infection were initially considered and later ruled out by biopsy and further investigations. The skin biopsy and bone marrow biopsy were not suggestive of LCH. Failure to thrive with absence of direct Coombs test (DCT) positive anti red blood cell (RBC) and antiplatelet antibodies ruled out Evan’s syndrome. After induction therapy for CMV infection, DNA copies of virus in the urine were not detected, so the possibility of unresponsive CMV infection was also excluded. The family took the child to a hospital closer home, where he continued to receive supportive care and transfusion therapy.

The child was readmitted about 42 weeks later to AIIMS with complaints of multiple abscesses, fever and worsening of skin lesions. Physical examination showed a febrile child with pallor and petechiae. He had severe thrombocytopenia (14 × 10⁹/L) and severe anaemia (Hb 5.7 g/dL). Serum urea, creatinine and liver transaminases were within normal limits. Blood culture grew methicillin-resistant Staphylococcus aureus (MRSA), and the patient was treated according to the sensitivity report. Immunodeficiency (primary and secondary) work-up was negative. At this time, the possibility of TA-GVHD was raised and investigated accordingly.

A sample of skin biopsy, hair follicle, buccal swab and peripheral blood of the child and his mother and father were collected for DNA profiling. Results of DNA profiling were positive for TA-GVHD.

DNA profiling

Result

The chimerism percentage in the patient’s different biological samples were estimated to confirm a diagnosis of TA-GVHD. This was suspected to be caused by a non-irradiated blood transfusion from the father to the patient, as the patient had received multiple transfusions from different donors, including his father, over the time period. Further observations revealed that the short tandem repeat (STR) profile of the patient had multiple peaks. The chimerism percentage was calculated with every possible combination: Type I, the total donor chimerism percentage was calculated by considering all the alleles present; Type II, those alleles which are common between patient and his father were considered as common; Type III, alleles which are common between patient and his father were considered as from the donor; Type IV, alleles from the father were excluded and alleles from other donors were used to calculate the donor chimerism percentage (Table 1). Allele bands shorter than 4 bp, corresponding to the main allele, were considered as stutter bands and were completely excluded from chimerism calculations.

OPEN IN VIEWER

Table 1.

Allelic designation of Marker D8S1179 for all four types used for chimerism analysis and percentage donor chimerism.

Specimen	Type I Alleles (% Ch)	Type II Alleles (% Ch)	Type III Alleles (% Ch)	Type IV Alleles (% Ch)
Blood	11,12,14,15,16 (81.5)	12*, 16 (68.71)	16 (80.53)	11,14,15 (45.16)
Buccal swab	11,12,14,15,16 (49.75)	12*, 16 (56.9)	16 (63.32)	11,14 (10.68)
Hair follicle	11,12,14 (58.88)	12*, 16 (54.64)	16 (65.35)	14 (5.74)

Donor chimerism was observed in all three biological samples (peripheral blood, buccal swabs and hair follicle) of the patient when analysed using different combinations of allele designation. Even when only paternal alleles were considered, a significant percentage of donor chimerism was observed, clearly indicating a typical case of TA-GVHD in a patient who had received non-irradiated blood from his biological father.

Discussion

TA-GVHD is a rare condition that responds poorly to immunosuppressive therapy and is often fatal. Current blood-bank practices in developed countries limit donor directed blood transfusions from relatives, and recommend the use of irradiated blood for all immune-compromised hosts, making it a rare complication of blood transfusion.^9,10 However, there are cases reported due to similar HLA antigens in immune-competent hosts ¹¹ and where family members, particularly first-degree relatives, have donated blood. In developing countries where most transfusions are family directed, it is important to consider this condition. Although rare, it can occur even with unrelated donors if they share the same HLA haplo types.

TA-GVHD and acute GVHD after BMT show some similar clinical symptoms. ⁶ GVHD after BMT does not show pancytopenia (decrease in two or more blood cell lines), since the bone marrow is not attacked through an immunologic mechanism because it is donor derived. TA-GVHD has a rapid fatal outcome due to infection or haemorrhage as a consequence of serious bone-marrow suppression. The other symptoms of diarrhoea and liver dysfunction are less common and are found in 40% of patients of TA-GVHD. ¹²

If a patient develops a skin rash and bone-marrow aplasia shortly after blood transfusion without any other illness, it is easy to suspect TA-GVHD. However, in a sick patient, the initial symptoms of TA-GVHD are often mistaken for infection, sepsis or drug effects. The current case was difficult, as the patient had had a diagnosis of CMV infection that could explain his low blood count. He had also received multiple blood transfusions over the period of his illness, both at AIIMS and at other health-care facilities. Since the illness was complicated by CMV infection and therapy with ganciclovir (which also causes pancytopenia), the exact point of the onset of illness is uncertain.

In 1989, the American Association of Blood Banks (AABB) issued recommendations related to the irradiation of all directed donor units from first-degree relatives. These guidelines clearly state that blood and blood components should be irradiated prior to transfusion to prevent the proliferation of certain types of T-lymphocytes that can inhibit the immune response and cause GVHD. This procedure is necessary for transfusion recipients at risk for GVHD, including fetuses receiving intrauterine transfusions and select immunocompetent or immunocomprised recipients.^13,14 Case reports of TA-GVHD in immmunocompetent patients have been reported. It was observed in these cases that the blood donor was homozygous for an HLA haplotype for which the recipient was heterozygous (one-way HLA matching). ¹³ Most of these cases have been observed in Japan and the Middle East.^15,16 According to the British Committee for Standards in Haematology (BCSH) guidelines, ¹⁷ ‘There is no indication for routine irradiation of cellular blood components for infants or children who are suffering from a common viral infection, who are HIV antibody positive, or who have AIDS. There is no need to irradiate red blood cells or platelets for children undergoing cardiac surgery unless clinical or laboratory features suggest a coexisting T-lymphocytes immunodeficiency syndrome’. However, the guidelines do state that neonates (a child younger than 1 month old) and all in utero transfusions must be irradiated. ¹⁷ Since the patient was not a neonate (<4 weeks old) and did not suffer from any known immunodeficiency, it was not recommended to give irradiated blood routinely. Direct blood donations are demanded by many families for cultural reasons, but it is a dangerous practice, unless this blood is irradiated. Unfortunately, many centres do not have facilities for irradiation. As per AABB guidelines, irradiation is required for ‘individuals receiving platelets selected for HLA or platelet compatibility, and individuals receiving units from blood relatives’.^13,14 Universal irradiation of all blood products is not practiced in any country, as irradiation shortens the shelf life of the blood components.

Gorman et al. showed that sharing HLA matching led to TA-GVHD even from an unrelated donor. Their results suggest that class I HLA haplotypes sharing is critical for the development of TA-GVHD where the patient recognises the transfused lymphocytes as belonging to the self while the transfused lymphocytes recognise the patient as foreign, thus initiating the disease. ¹⁸

Another cause of TA-GVHD in immunocompetent patients explained by Petz et al. ¹¹ is the transfusion of major histocompatibility complex (MHC homozygous) in parents to F1 hybrid (MHC heterozygous) offspring.¹⁹ The donor lymphocytes can escape detection and react with MHC markers encoded by the recipient’s other chromosome. In another case, an immunocompetent male patient received blood from 22 unrelated donors followed by open-heart surgery. Two of the donors were homozygous for one of the patient’s HLA class I haplotypes. ²⁰ Similarly, Naveen et al. have reported a case of a two-year-old immunocompetent girl who received whole blood from her biological father and developed TA-GVHD. ²¹

STR analysis for the diagnosis of TA-GVHD is well established. ²² The presence of donor alleles in post-transfusion patient samples that were not found in pre-transfusion samples strongly supports the presence of chimerism. Previously, it was believed that pre-transfusion DNA could be obtained from a patient’s hair follicle. ²³ However, in the present case, TA-GVHD was diagnosed in an immunocompetent patient who received blood transfusions from multiple donors. The patient’s hair follicle sample also revealed the donor profile. Proper care was taken to eliminate possible contamination of the hair follicle with hematic cells. Five or six hair follicles were used for analysis, which were washed properly with saline prior to isolation. The possible mechanism for the presence of donor-derived cells in the patient’s hair follicles is the de-differentiation of the cells. This patient was diagnosed with TA-GVHD through association of the characterised clinical manifestation (including skin rashes, fever and cytopenia). The implicated transfusion and the onset of symptoms were consistent with the diagnosis. Usually in newborn babies, symptoms of TA-GVHD arise 4–30 days after transfusion as a result of engraftment in the recipient of viable transfused lymphocytes from cellular blood products. ²⁴ It is typically seen in individuals with profound defects of cell-mediated immunity, but on rare occasions it is seen in apparently immunocompetent individuals. The symptoms of TA-GVHD in the present patient may have been missed due to concomitant illness and directly inquiring about donor-directed blood transfusion, which must be done. Analysis of the hair follicles showed mixed chimerism. This has already been observed in hair follicle samples after hematopoietic stem-cell transplantation. ²⁵ To the best of the authors’ knowledge, this is the first case reporting chimerism in the hair follicles of a TA-GVHD patient.

STR markers helped in the early diagnosis of TA-GVHD in this case, but this patient succumbed to his illness. The major utility of this study is in medico-legal cases, since blood and buccal swabs may have degraded and may not be readily available sources of DNA. In such a scenario, the hair is of great significance because it can be easily used.