Nadir B cell counts are significantly correlated with the risk of Kaposi's sarcoma

Abstract

Infection with HIV-1 is known to impair B cell function. To further elucidate the role of B cells during infection and tumorigenesis, we studied their numbers in cases of AIDS-related Kaposi's sarcoma (KS) during the HAART era. Patients with AIDS-related KS were identified from a database of 4,480 HIV-1 positive individuals and the incidence of KS and rate ratio was stratified according to nadir number of B cells, measured as the CD19 count. In an unadjusted model, we observed that lower B cell counts were associated with a statistically significant increased risk of KS development (p < 0.001). We also observed a trend toward increased counts during KS resolution. When adjusted for nadir CD4 count in a multi-variable model, higher B cell counts were protective against KS development (p = 0.015). These data highlight a potential role for B cells and therefore the humoral immune system in KS aetiopathogenesis. © 2003 Wiley-Liss, Inc.

There are a wide range of B cell abnormalities observed during infection with HIV-1, the first of which was described soon after HIV was identified as the causative agent of AIDS, 2 decades ago.1 The causes of B cell dysfunction include HIV-1 induced hyperactivation manifesting as hypergammaglobulinemia,1, 2, 3 terminal differentiation4, 5 and increased susceptibility to apoptosis.6, 7 It is known that that B cells of HIV-viremic patients respond poorly to mitogenic and antigenic stimuli in vitro1, 8 and have poor antibody responses to both T cell-dependent and independent antigens.9, 10, 11 Ineffective B cell costimulatory function in HIV-viremic patients has been observed recently, associated with low induction of expression of the CD28 ligands, CD80 and CD86, on B cells.12 In addition, these adaptive immune cells have also been recently described to have an ability to express CD25 in response to activated CD4+ cells, resulting in an inability to mount proliferative responses to interleukin-2.13

We investigated whether changes in B cell number were related to the risk of developing an AIDS-defining illness, Kaposi's sarcoma (KS), in our cohort of HIV-1 infected individuals. B cell counts are routinely measured in this department using flow cytometry (Beckman Coulter, Oxford, UK) with antibodies to CD19 (BD Biosciences, Oxford, UK). B cell counts, defined as the CD19 count, were linked to 4,480 individuals from the Chelsea and Westminster HIV-1 cohort (1996–2003), one of the largest single centres in Europe. The nadir B cell count was defined as the lowest ever CD19 count recorded during patient follow up to the time of KS diagnosis or last follow up. Person days of follow up were converted to person years at risk (PYAR). To keep the coefficient of the PYAR constant, this was log transformed and used as the offset in the Poisson regression. The data were analysed using the Genmod procedure in SAS version 8.0 with log link and Poisson error distributions. PYAR was estimated from entry into the cohort to either end of study period, the development of KS, the last recorded visit or if the patient had died during their follow up then their death date, as for a study published recently.14

We observed a statistically significant relationship between nadir B cell count and KS risk (p < 0.001; Table I). Those HIV-1 infected individuals with a B cell count of <39 cells/mm had a rate ratio of developing KS of 1.98 (95% confidence intervals [CI] = 1.32–2.97). As the nadir CD19 count decreases, a significant increase in the incidence of KS was observed.

Table I. Unadjusted Log Linear Model Showing Rate Ratio of KS Incidence during the HAART era1

Variable	HIV patients at risk during the HAART era (n = 4,480)	Kaposi's Sarcoma cases n (%) (n = 198)	Incidence of KS per 1,000 patient years	Rate ratio	95% CI	p-value
1 B cell counts were divided into quartiles. 2 If KS then year of KS diagnosis. 3 Reference category. 4 Adjusted for nadir CD4 count.
Year of censoring from cohort2
1996	415	65 (15.7)	7.6	13	—	<0.001
1997	366	49 (13.4)	6.1	0.81	(0.56–1.18)
1998	349	27 (7.7)	3.8	0.50	(0.32–0.78)
1999	326	34 (10.4)	4.8	0.63	(0.42–0.95)
2000	534	21 (3.9)	1.9	0.25	(0.15–0.40)
2001	2,490	2 (0.1)	0.03	0.00	(0.00–0.02)
Nadir CD19
Missing	581	9 (1.6)	1.2	0.79	(0.38–1.67)	<0.001
<39	962	81 (8.4)	2.9	1.98	(1.32–2.97)
39–76	979	49 (5.0)	1.9	1.28	(0.83–2.00)
77–130	981	26 (2.7)	1.0	0.70	(0.42–1.18)
≥131	977	33 (3.4)	1.5	13	—
Adjusted nadir CD194
Missing	581	9 (1.6)	13.03	1.66	(1.30–2.12)	<0.001
<39	962	81 (8.4)	8.86	1.13	(0.92–1.37)	0.223
39–76	979	49 (5.0)	8.15	1.04	(0.85–1.26)	0.710
77–130	981	26 (2.7)	6.05	0.77	(0.62–0.95)	0.015
≥131	977	33 (3.4)	7.85	13

Because a low CD4 cell count is an independent risk factor in the development of KS,14 we adjusted the CD19 counts for nadir CD4 count. When corrected, CD19 counts <76 cells/mm³ were not associated with changes in risk of developing KS. Counts greater than 76 cells/mm³ were protective against KS development (rate ratio 0.77, 95% CI = 0.62–0.95, p = 0.015, Table I; note that this is against the reference category of >131 cells/mm³).

We then followed patients with AIDS-related KS to observe changes in B cell count over time. In 97 individuals with KS resolution during the HAART era, we observed an increase in B cell count over 1 year, which was not statistically significant due to the wide interquartile range (Fig. 1).

Figure 1. Changes in B cell count during KS resolution in 97 individuals with AIDS-related KS. The x-axis demonstrates months since KS occurrence. Median and interquartile range of nadir CD19 counts is shown.

In the study by Moir et al.13 the authors observed a profound improvement in B cell dysfunction in those individuals whose plasma viremia was effectively suppressed by anti-retrovirals. We observe that a low B cell count predisposes individuals with HIV-1 infection to the development of KS and that high CD19 counts appeared protective. We observe a trend toward increased B cell counts during KS resolution, a feature not seen in a similar study of natural killer counts.18 Although both cytotoxic T lymphocyte15, 16 and natural killer17 responses have been shown to be mechanisms that contribute to the resolution of KS, our study also points to the humoral arm of the adaptive immune system as having a significant role in the protection against KS.