To Determine the Histomorphological Variants of Kaposi’s Sarcoma in HIV-infected Patients in the University of Benin Teaching Hospital

Abstract

Introduction:

There are different histopathological variants of Kaposi’s sarcoma (KS) aside the typical patch, plaque, and nodular variants. The baseline data of the unfamiliar variants of this lesion have not been determined in our environment. Their determination will bring to the foreclose differential diagnoses that may lead to diagnostic pitfalls.

Aim and Objectives:

To determine the histomorphological variants of KS in the University of Benin Teaching Hospital over a 3-year 2-month period from May 1, 2014, to July 31, 2017.

Materials and Methods:

This was a prospective study. The patients were all adult human immunodeficiency virus-positive highly active antiretroviral therapy users who had symptomatic cutaneous lesions that were histopathologically identified as KS. The histopathological variants of KS were noted. The Statistical Package for Social Sciences, version 16, was used to analyse the data.

Results:

Fifty patients had KS. The histomorphological variants (stages) of KS, as seen in this study, ranged from the typical progressive lesions of patch (13.2%), plaque (43.4%), and nodules (18.4%) to rarely diagnosed variants that are lymphangioma-like KS (6.6%), lymphangiectatic KS (5.2%), hyperkeratotic KS (1.3%), keloidal KS (9.2%), ecchymotic KS (1.3%), and pyogenic granuloma-like KS (1.3%). The mixed pattern (spindle cells and vascular channels) was the predominant (63%) histologic cellular group in this study.

Conclusion:

In addition to the common histopathological subtypes of KS, this study also found less common histopathological variants.

Introduction

Skin lesions are the typical presentation of Kaposi’s sarcoma (KS), a malignant vascular endothelial tumour that is locally aggressive but rarely metastasises.[1] Idiopathic multiple-pigmented sarcoma was the original name for it.[2] There have been four clinical and epidemiological types which are identified, including but not restricted to KS related with acquired immunodeficiency syndrome (AIDS).[3,4,5,6,7]

KS has a range of histomorphological variants aside from its usual variants of patch, plaque, and nodular variants.[8]

The initial lesion is typically observed clinically to grow gradually from patches to plaques and then to nodules with or without ulceration and bleeding.[8,9] Anaplastic, lymphedematous, and telangiectatic subtypes are morphologic variations that can be discovered in earlier literature.[8] Regardless of the subtypes, the lymphedematous form may present clinically with a bulla-like appearance.[8] The subgroups in this case are bullous KS, lymphangiectatic KS, and lymphangioma-like KS.[8] The recently identified forms of KS include ecchymotic, keloidal, micronodular, pyogenic granuloma-like sarcoma, and hyperkeratosis (verrucous).[8] Regressing KS and KS exacerbation are variations that are related to therapy.[8]

Histopathological study is considered the gold standard for the diagnosis of KS, and early diagnosis and commencement of treatment would help to improve the prognosis.[10,11]

There are hardly any studies on the histopathological subtypes of KS in this setting in particular, in Nigeria, and the sub-Saharan region in general. To characterise the histomorphological variants of KS lesions at the University of Benin Teaching Hospital (UBTH), this study was conducted. By doing so, it generated the database of histomorphological variants of this lesion in our environment.

Materials and Methods

This study was a descriptive observational study that was carried out at the UBTH, Benin City, Edo State in the South–South geo-political region of Nigeria over 3 years, 2 months period from May 1, 2014, to July 31, 2017.

The minimum sample size for this study was determined using the Cochrane formula below.[12]

n = Minimum sample size

Z = Normal standard deviate 95% confidence interval (Z = 1.96)

q = 1−P

d = Margin of error (5%)

P = Proportion of the variable of interest

The prevalence value of 2% from a study done by Onyemelukwe and Musa[12] (Zaria) was used.

The sample size obtained from the formula was 30. The non-response rate was 10% = 3. The minimum sample size was therefore 33.

The minimum calculated sample size was 33 cases of human immunodeficiency virus (HIV)-positive patients >18 years of age. These patients were either attendees of the out-patient clinic or were on admission during the study period. They were examined clinically for the presence of lesions suspected to be KS. A consecutive sample technique was used in this study. All HIV patients with lesions suspected to be KS at presentation in the clinics and wards were interviewed and examined, after obtaining their consent. The clinically suspected lesions were biopsied, and the specimen was sent to the Department of Anatomic Pathology Laboratory for routine tissue processing using Leica automatic tissue processor, and this was followed by haemaetoxylin and eosin staining. The slides were reported by a pathologist dedicated to this study. The patients with histopathological diagnosis of KS were recruited into the study.

The inclusion criteria were all HIV-positive patients with lesions clinically suspected to be KS whose biopsies were confirmed by the histopathologist to be KS.

The exclusion criteria were patients with other immune-compromising diseases (such as diabetes mellitus, lymphomas, and chronic renal disease), patients on immunosuppressants (such as steroids or cytotoxic drugs), and pregnant women.

The paucity of similar research work to compare the findings of this research work with and the absence of immunohistochemical studies for HHV8, CD31, and CD34 are limitations of this study.

Results

Fifty patients were recruited into this study after histological confirmation of KS. There were 28 males (56.0%) and 22 females (44.0%). The mean age of these patients was 40.4 ± 11.5 years. Male patients were older with a mean age of 44.6 ± 12.3 years compared to female patients whose mean age was 35.2 ± 6.8 years.

The participants in this study provided 76 biopsies in total. The histomorphological variants of KS that were seen in this investigation ranged from the conventional histopathological subtypes (patch, plaque, and nodular) to the less common histopathological variants [Table 1a]. The conventional histopathological subtypes (75% of biopsies) were more common than the less common histopathological variants (25% of biopsies). The most common variants were the plaques of the conventional histopathological subtypes (43.4% of biopsies), while the least common variants were jointly hyperkeratotic, ecchymotic, and pyogenic granuloma-like KS accounting for 1.3% of biopsies clinically suspected to be KS.

Table 1a.

Histomorphological variants of Kaposi’s sarcoma (KS) and their relative frequency

Histomophological variants	Frequency	Percentage (%)
(A) Conventional histopathological subtypes
Patches	10.0	13.2
Plaques	33.0	43.4
Nodules	14.0	18.4
Subtotal	57.0	75.0
(B) Less common histopathological variants
Lymphedematous
(I) LLKS	5.0	6.6
(II) Lymphangiectatic KS	4.0	5.2
Hyperkeratotic KS	1.0	1.3
Keloidal	7.0	9.2
Ecchymotic	1.0	1.3
Pyogenic granuloma-like KS	1.0	1.3
Subtotal	19.0	24.9
Grand total	76.0	100.0

KS: Kaposi’s sarcoma, LLKS: Lymphangioma-like Kaposi’s sarcoma

The most frequent histologic features were the presence of spindled-shaped cells in 79% of cases. Table 1b shows the distribution of other histologic features of KS.

Table 1b.

Histologic feature of Kaposi’s sarcoma

Histologic features	Frequency	Percentage (%)
Slit-like vessels	58	73
Inflammatory cells	39	51
Spindle-shaped cells	60	79
Atypical spindle-shaped cells	5	7
Haemosiderin deposits	36	47
Hyaline globules	42	55
Extravasated red blood cells	48	63
Promontory sign	15	20
Irregular ecstatic anastomosing vascular channels	5	7
Dilated thin-walled lymphatic vessels intra- and peri-tumoural	4	5
Epidermal collarette surrounding KS lesional cells	1	1.3
Marked red blood cell extravasation	1	1.3
Broad bands of perilesional collagen fibres	7	9
Epidermal acanthosis and hyperkeratosis that overlies a fibrotic superficial dermis and the KS lesion located deep into the dermis	1	1.3

Inflammatory cells = lymphocytes and plasma cells

The mixed group consisting of vascular channels and spindle-shaped cells is the most common cellular pattern of the histologic features, as shown in Table 1c. See the same Table 1c for other cellular patterns and their relative frequency.

Table 1c.

Histologic features grouped according to cellular pattern

Cellular pattern	Frequency	Percentage (%)
Mixed group	48	63
Spindle cell predominant group	23	30
Anaplastic group	5	7
Total	76	100

The photomicrograph of the histologic sections of some of these lesions is seen in Figure 1a–n.

Figure 1.

Photomicrographs of Kaposi’s sarcoma (KS). (a) Patch variant of KS showing slit-like spaces between collagen bundles and extravasated red blood cells. The basal layer shows hyperpigmentation. (b) Plaque variant of KS showing spindle cell proliferation arranged in short fascicles and proliferation of blood vessels. (c) Nodular variant showing well-defined nodules of spindle-shaped cells and extravasated red blood cells. Haematoxylin and eosin. (a) ×100 magnification, (b) ×400 magnification (c) ×100 magnification. (d and e) Are hyperkeratotic variants showing epidermal acanthosis and hyperkeratosis that overlies a fibrotic superficial dermis in a patient with elephantiasis nostra verrucosa. The KS lesion is not shown because it is located very deep into the dermis. (d) ×40 and (e) ×400 magnification; (f) is lymphangioma-like KS showing dilated anastomosing vascular channels and promontory sign ×100 magnification; (g) is lymphangiectactic KS showing dilated thin-walled lymphatic vessels intra- and peri-tumoural (single arrow for intratumoral and double arrows for peri-tumoural vessel, ×100 magnification). (h) is ecchymotic KS showing marked red blood cell extravasation, ×400 magnification; (i and j) are ×40 and ×400 magnification of pyogenic granuloma-like KS showing the presence of epidermal collarette and promontory sign; (k) is keloidal KS showing collagen flanking tumoural cells. (l–n) show the mixed pattern, spindle cell predominant pattern, and anaplastic pattern, respectively. The spindle cells in n are very cellular and pleomorphic, and there is the presence of abnormal mitosis. (l) ×100 magnification; (m and n) ×400 magnification

Discussion

This study reported rare variants of KS aside from the conventional patch, plaques, and nodular variants. This study also brought to the fore the individual frequencies of these variants relative to each other. Unlike this study, the histomorphological variants of KS were not discussed in the earlier studies on AIDS-associated KS (AAKS) conducted in Nigeria.[13,14,15,16,17] A later study by Akinde et al. from Lagos reported that the plaque variant of the conventional histopathological subtypes was the most frequently encountered variant in their study; the individual frequencies of the other variants of AAKS were not reported by Akinde et al.,[18] thus limiting the comparison of the findings of both studies with respect to histopathological variants. A more recent study by Menkiti et al.[19] from Nnewi revealed that the variants of KS ranged from the conventional histopathological subtypes to the less common histopathological variants, however, their frequencies were not stated. On the one hand, this has limited the comparison between the study from Nnewi by Menkiti et al.,[19] while on the other hand, the study from Nnewi has given insight into the existence of less common variants of KS in Nigeria, a finding that is consistent with this research work. The existence of rare variants of KS in Nigeria needs to be emphasised because previous studies have documented that most pathologists are not well-acquainted with the less common variants of this lesion.[8,18,19] This, they opined, may be a basis for misdiagnosis of cutaneous KS that ought to be guided against.[8,18,19] In the light of this, Akinde et al.[18] bemoaned the low level of childhood KS in their study in particular and Nigeria in general. This is partly attributed to a possibility of misdiagnosis of paediatrics KS with vascular malformation differentials which include but are not limited to haemangioma and granuloma pyogenicum.[18] As far as our literature search could go on the internet, there is a paucity of data on similar work from within Nigeria, West Africa, and Africa, and globally that would have ensured a more robust discussion. Perhaps the rarity of some of these variants, as documented by the World Health Organisation,[20] and the low awareness as documented by other studies,[8,18,19] may be responsible for the relative paucity of studies on the histomorphological variants of KS in recent times.

It was found that the cellular pattern of the histologic features in this research work was comparable to those from earlier investigations conducted in this environment by Forae and Obaseki.[21] According to their findings, 50.6%, 23%, and 16% of the cases were accounted for by the mixed pattern, monomorphic or spindle cell dominating pattern, and anaplastic pattern, respectively.[21] This result agrees with the findings of O’Connel[22] and Mandong et al.[23] Mixed, monomorphic, and anaplastic cellular patterns made up 61.5%, 20%, and 19.5%, respectively, of the research population, according to Mandong et al.[23] In a similar vein, O’Connel[22] reported that the mixed, monomorphic, and anaplastic cellular patterns accounted for 80%, 16%, and 4%, respectively, of the proportion of their study population used for calculating the cellular pattern (i.e., 147 subjects were used to calculate the cellular pattern out of the 159 subjects enrolled for the study).

From the foregoing, the results of this study will expand the body of knowledge on the histomorphological variants of KS in our local area. Our study has shown the large range of histomorphological variants of KS which would help in the long run with the rapid histopathological diagnosis of this lesion. The overall goal would be a possible decrease in the number of KS cases that might be misdiagnosed, stimulate the interest of other researchers to determine the scope of histomorphological variants of KS in their locality with the concurrent determination of their relative frequencies to one another and serve as a basis for comparison with future researches within our environment and beyond.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.