Blood group and Human Leucocyte Antigen sub-type as determinants to keloid formation and Recurrence in Keloid Patients; A prospective longitudinal cohort study

The role of genetic factors in keloid is a rmed by the fact that keloids have been shown to occur among members of the same family.. We undertook this study to determine whether there is any association between patients’ bloodgroup and HLA sub-types to keloids and keloid recurrence. This was a prospective longitudinal study of patients with keloids and a control of patients managed for other surgical conditions with no keloids. Blood was taken from each patient and analysed for blood group and HLA sub-types using the sequence specic primer geno-typing. Data captured were summarized and analysed using students T-test to compare means. Probability values signicance was at 0.05. A total of 90 patients with keloids and 59 in a control group were followed up in the study. The male to female ratio of the patients was 2:1. The most common blood group for both groups was blood group O at 51.3% and 49.2%, followed by blood group A and B respectively. Patients with keloids had a signicantly higher positive alleles with DQA*01, DQB1*05, DQB1*06 and DRB1*15. There was an association between blood group A and DQB1*06 and recurrence. In conclusion, this study demonstrates that there is signicant difference in HLA sub-types among patients who form keloids and the non-keloid forming patients among our study cohorts. Salient differences were also noted in patients with keloid recurrence based on their blood group, a pointer to the signicance of genetic factors in keloid pathogenesis and severity.


Introduction
Pathogenesis of keloid disease is still not well established. Keloid occurrence has generally been classi ed as sporadic, in individuals with no known members of the family with keloids or familial where members of a given family have positive history of keloids. Familial keloids have been shown to contribute to about 50 per cent of keloid patients indicating a strong inheritance pattern in the pathophysiology of the disease [1].A number of pedigree studies have demonstrated keloids to either have an autosomal or recessive mode of inheritance [2][3]. Other studies have suggested X linked mode of inheritance [4]. HLA studies have demonstrated some consistentassociation with some sub-types and keloid formation especially HLA DRB*16 sub-type [5]. Majority of these studies have however been done in populations with low keloid prevalence with non in Africa, a keloid endemic zone.An association between ABO blood groupingand keloids have also been documented by a number of authors [6]. Keloids have high recurrence rates irrespective of the treatment modalities provided [7][8]. Whether a patient's genetic factors in uence keloid recurrence has not been demonstrated. We undertook this study to demonstrate any signi cant differences between various blood groups andHLA sub-types in patients with keloids and those without and to determine whether they could in uence keloid recurrence.

Materials And Methods
Patients with keloids were involved in the design of the study through pretested questionnaires to determine the most common clinical features of keloid recurrence.All methods were carried out in Page 3/10 accordance with relevant guidelines and regulations. The study was approved by the local ethics and research committee. Informed consent was sort from all patients to participate in the study. For the minors informed consent was taken from the parents or legally acceptable representatives . Patients with keloids managed by surgical excision and a control with no keloids were followed up for a minimum duration of one year. History and physical examination was taken. Pain and pruritus scores were done using the respective visual analogue scores.This was repeated at regular intervals during the follow up period. The control group was of patients admitted for elective surgical procedures with no keloids nor family history of keloids. They were followed up for one year to determine if they would also formkeloids.
From each patient, blood was taken and analysed for blood group and HLA sub-types.HLA typing was done using sequence speci c primer geno-typing. All patients with keloids had surgical excisionby a senior plastic surgeon. Peri-operative management of the patients was similar.Post-surgery all patientswith keloids weresubjected to super cial radiotherapy within 24 hours of surgery.Recurrence was de ned as keloid regrowth or worsening pain and or pruritus requiring medical intervention. Data captured were summarized and analysed using students T-test to compare means. Probability values signi cance was at 0.05.

Results
A total of 90 patients with 104 keloids and a control group of 59 patients were followed up during the study. The male to female ratiowas 1:2 with 60 females and 30 male patients.For the control group the male to female ratio was 1:2 with 39 female and 20 male patients.The age rangewas 15 to 65 years with a mean age of 29.5 and a median age of 20 -25 years. The age range for control patients was 15.5 to 64 years with mean age of 29.7 years and a median ageof 20-25 years. There was no statistical signi cance difference between the keloids group and the control population ( Table 1). The anatomical location of the keloids were as follows; ears (n=47), abdomen (n=10), cheek (n=15), upper limb (n=5), back (n=13), chest and neck (n=13) and scalp (n=1).All 90 patients and control had their blood taken for blood groups while 80 of the 90 patients had their blood taken for HLA studies. Eighteen patients of the 80 had keloid recurrence (KR) while 62 did not have any form of recurrence (NKR).
The most common blood group for patients with Keloidsand the control group with no keloids was blood group O at 51.3% and 49.2% (p-value 0.552), followed by blood group A and B respectively. There was no statistical signi cance difference in the blood groups of patients who formed keloids and the control group(Table2).
Comparison between blood groups in patients with KR and NKR patients demonstrated that keloid patients with blood group A were more prone to recurrence compared to the other blood groups (p-value <0.011) while patient with blood group B were list likely to have keloid recurrence(p-value <0.001) (Table3).
Out of 90 patients with keloids who were followed up in the study 80 patients had their blood analysed for HLA. Of the 80patients 18 had keloid recurrence (KR)and 62 had no recurrence (NKR).Comparison of the HLA sub-types between the control group and patients who had keloids showed signi cance differences in the alleles of HLADQA1*01, DQB1*05, DQB1*06, and DRB1*15 all in favour of patients who had keloids (Table4). Analysis of HLA sub-types between patients who had keloids recurrence (KR) and those without (NKR), revealed DQBI*06 to be signi cantly elevated in the keloid recurrent group(Table5).

Discussion
The in uence of genetic factors in keloid formation could be extrapolated from the facts that keloids seem to portray a distinct pattern of inheritance and a strong familial tendency [1][2][3]. Though not unique to keloids, genetic factor have been shown to have an in uence on various disease processes whether infective benign or malignant origin. Most studied and easily applied genetic variants have been the ABO and the HLA systems probably due to the fact that they are routinely used during blood transfusions aswell as tissue typing and organ transplants. Our study noted signi cant associations between this two genetic systems of the body and keloid patients.
The role and in uence of ABO blood groups in the pathogenesis and outcome of keloid management has been established before. Ramma kishnan et al in India found blood group A to be signi cantly higher in patients with keloids than in the normal population [9].These ndings were similar to those of Shaheen et al. in a Syrian study recently who found blood group A to be signi cantly more common in patients with keloids than in the normal cohort [6]. However, Abbas Moure Toure in Togo, like in our study demonstrated no statistical signi cance difference with blood groups in patents with keloids and a control population of patients who presented with other dermatological conditions [10]. Our study on the other hand found keloid patients with blood group A to be more prone to keloid recurrence compared to the other blood group with blood group B being less prone to recurrence.
Human leucocyte Antigen (HLA), key genes responsible for the development of the cellular immune system, have been shown to play a critical role in not only autoimmune conditions but also in benign and malignant conditions.Mutations on some of the alleles are thought to be responsible for the development of a number of autoimmunedisorders such as celiac disease, anky losing, spodylisthesis and rheumatoid arthritis [11][12]. Even closer to keloids Sclerotic skin conditions such as sarcoidosis and systemic sclerosis have also been shown to have associations with particular alleles [13].Though keloids disease are not classi ed as autoimmune disease several studies have shown an association between some alleles and the disease bringing to the fore possibilities that the immune system could play a critical role in keloid formation.Wen-Sheng Lu found DQB1*0501, B*07-DRB1*15, DQB1*0503-DRB1*15 (P<0.05) to be associated with keloid formation among Chinese Ham patients cohorts [14]. Another study by Brownet alfound an association between HLA DRB1*15 and keloid formation in apatient cohort of Caucasians [15]. However, in a study in the Caribbean with Brown et alno association was demonstrated between HLA DRB*15 alleles and keloid patients of African descent [16]. Inour study we demonstrated HLA alleles DQA*01, DQB1*05, DQB1*06 and DRB1*15 to be signi cantly associated with keloid patients. Even further Allele DQB1*06 were found to be signi cantly associated with keloid recurrence.
Our study ndings seems to add more evidence to the fact that keloid pathogenesis and recurrence seem to be in uenced by the individual's immune system. The exact mechanism on how this is achieved is still not clear and probably requires mores studies. It is however postulated that certain alleles such as DRB1*15 may be more e cient at antigen binding and presentation to the antigen presenting cells following trauma or any injury leading to a stronger in ammatory and proliferative phase of wound healing resulting in massive collagen deposition akin to keloid formation [13]. Evidence of the strong in ammatory response in keloid patients include the fact that keloid patients in a number of studies have been shown to have elevatedpro-in ammatory cytokines compared to the normal population suggesting a pro-in ammatory state of the body [17][18][19]. Majority of keloid specimens also have been shown to have a high concentration of in ammatory cells such as macrophages, lymphoctyes and mast cells [20].
Further normal bro-blasts in vitro subjected to in ammatory cytokines such as interleukin 6 have responded by massive synthesis of collagen bres. These activities were stopped by antibodies against interleukin 6 implying that abnormal bro-blast activity in keloid patient is secondary to in ammatory response by cytokines [18,21].

Conclusion
Keloid patho-physiology from our study ndings seem to be in uenced by the genetic composition of the patients. Patients with blood group A are more prone to keloid recurrence a possible indication of the disease severity in this group of patients. While a number of HLA sub-types were identi ed to be signi cantly associated with keloid disease in our cohort, HLA DRB1*15 has also been noted in several other studies from different patient population to be associated more with the disease. There is need to carry out more research on this particular loci on its role on keloid formation. The association between several type 2 alleles and keloid formation in our study and other studies strongly suggest an immunological aspect in keloid formation.

Declarations
Funding statement: This work was supported by Foundation National institute of Health grant number K01AR071479 Con ict of interest . There is no con ict of interest to declare Availability of Data/Material . The data for this study will be availed at the University of Nairobi repository