Table 1 summarizes the baseline characteristics of HIV-1-positive women diagnosed in Israel in the years 2010-2012, 2013-2015 and 2016-2018. Median age at diagnosis was 38. Main route of HIV transmission (82.4%, n=629) was sexual contact; only 10.9% (n=83) were injecting drug users (IDUs). Most women were immigrants: 41.2% (n=314) were born in the FSU, 32.2% (n=246) in SSA and only 11.4% (n=87) were born in Israel. While the total number of women identified remained stable over the study period, a significant yearly decline in the proportion of SSA immigrants versus a constant increase in women originating from the FSU was observed (p<0.001). Similarly, while the overall prevalence of subtype C (41.8%, 141/337) and A (38.6%, 130/337) diagnosis was similar, the later years of the study were associated with a decline in the number of subtype C carriers and an increase in the number of subtype A carriers.
Table 1: Characteristics of women diagnosed with HIV, Israel, 2010-2018
|
All years (2010-2018) N=763
|
2010-2012 N=246
|
2013-2015 N=257
|
2016-2018 N=260
|
p value
|
Median age at diagnosis (IQR)
|
38 (31-46)
|
37 (29-43)
|
37 (30-46)
|
40 (34-48)
|
<0.001
|
Place of birth, n (%)
|
|
|
|
|
|
SSA
|
246 (32.2)
|
110 (44.7)
|
64 (24.9)
|
72 (27.7)
|
|
FSU
|
314 (41.2)
|
77 (31.3)
|
107 (41.6)
|
130 (50)
|
|
Israel
|
87 (11.4)
|
30 (12.2)
|
32 (12.5)
|
25 (9.6)
|
<0.001
|
Other/Unknown
|
116 (15.2)
|
29 (11.8)
|
54 (21)
|
33 (12.7)
|
|
Risk Groups, n (%)
|
|
|
|
|
|
Sexual contact
|
629 (82.4)
|
202 (82.1)
|
211 (82.1)
|
216 (83.1)
|
|
IDU
|
83 (10.9)
|
34 (13.8)
|
33 (12.8)
|
16 (6.2)
|
0.001
|
Other/Unknown
|
51 (6.7)
|
10 (4.1)
|
13 (5.1)
|
28 (10.8)
|
|
HIV-1 Subtype (N)
|
337
|
123
|
109
|
105
|
|
A, n (%)
|
130 (38.6)
|
45 (36.6)
|
40 (36.7)
|
45 (42.9)
|
|
B, n (%)
|
34 (10.1)
|
10 (8.1)
|
17 (15.6)
|
7 (6.7)
|
|
C, n (%)
|
141 (41.8)
|
61 (49.6)
|
44 (40.4)
|
36 (34.3)
|
0.014
|
G/AG, n (%)
|
23 (6.8)
|
3 (2.4)
|
7 (6.4)
|
13 (12.4)
|
|
Other, n (%)
|
9 (2.7)
|
4 (3.3)
|
1 (0.9)
|
4 (3.8)
|
|
CD4 (cells/mm*3) (n=171), median (IQR)
|
263 (121-466)
|
285 (146-496)
|
270 (133-492)
|
234 (73-394)
|
0.283
|
VL (Log c/mL) (n=236), median (IQR)
|
4.5 (3.9-5.4)
|
4.4 (3.7-5.4)
|
4.6 (4.1-5.2)
|
4.8 (4.3-5.3)
|
0.519
|
Data are presented as n (%) or median (IQR); IQR- interquartile range; VL-viral load; SSA- Sub-Saharan Africa; FSU-former Soviet Union; IDU-injecting drug users
A comparison of the characteristics of women born in SSA, FSU, Israel or elsewhere (Table 2) showed that most women from the FSU (79.7%) were carriers of subtype A, while 90.3% of those from SSA carried subtype C (p<0.001). Women were diagnosed with low (<350 cells/mm*3) CD4+ cell counts (Table 1), with lower median counts among women immigrating from SSA and the FSU (246 cells/ mm*3 and 262 cells/mm*3, respectively) as compared to Israeli-born women (391 cells/mm*3, p=0.042, Table 2).
Table 2: Characteristics of women diagnosed with HIV in 2010-2018, by place of birth
|
SSA
|
FSU
|
Israel
|
Other
|
p value
|
HIV-1 Subtype (N=337)
|
n=124
|
n=123
|
n=43
|
n=47
|
|
A, n (%)
|
2 (1.6)
|
98 (79.7)
|
14 (32.6)
|
16 (34)
|
<0.001
|
C, n (%)
|
112 (90.3)
|
8 (6.5)
|
4 (9.3)
|
17 (36.2)
|
Non A , C, n (%)
|
10 (8.1)
|
17 (13.8)
|
25 (58.1)
|
14 (29.8)
|
CD4 (cells/ mm*3, N=159)
|
n=52
|
n=81
|
n=26
|
No data
|
|
Median (IQR)
|
246 (99-348)
|
262 (106-488)
|
391 (179-738)
|
|
0.042
|
TDRM by class (N=337)
|
|
|
|
|
|
All TDRM, n (%)
|
16 (12.9)
|
10 (8.1)
|
7 (16.3)
|
2 (4.3)
|
0.170
|
NNRTI, n (%)
|
11 (8.9)
|
6 (4.9)
|
7 (16.3)
|
0
|
0.014
|
NRTI n (%)
|
5 (4)
|
4 (3.3)
|
0
|
1 (2.1)
|
0.582
|
Data are presented as n (%) or median (IQR-interquartile range); Significance for differences was measured using chi-squared test, Fisher's Exact test, or Kruskal-Wallis test. TDRMs-transmitted drug resistance mutations; NNRTI-non nucleoside reverse transcriptase resistance mutations; NRTI-nucleaotide reverse transcriptase resistance mutations
Resistance analysis revealed that 10.4% (35/377) of women carried viruses with resistance mutations, with 7.1%, 3%, and 1.8% of women carrying NNRTI, NRTI and PI TDRM, respectively. While the proportion of women with NNRTI TDRM increased significantly (p=0.017) between 2010-2012 and 2016-2018, paralleling a non-statistically significant increase in the overall prevalence of women with any HIV-TDRM diagnosed in these years, the rates of women with NRTI and PI TDRM remained stable. Moreover, in 2016-2018, no women with PI TDRM were identified (Figure 1). All these results were further corroborated by Poisson segmented regression. No significant differences was observed in the prevalence of women with any TDRM between the different birth-places (p=0.170). Interestingly, the proportion of native Israeli women born in carrying a NNRTI TDRM virus (16.3%, p=0.014) was significantly higher compared its prevalence among women born in other countries (Table 2).
Figure 1: Prevalence of women diagnosed with HIV-1 TDRM in 2010-2012, 2013-2015 and 2016-2018. Prevalence of women with any TDRM (total) and with NRTI, NNRTI and PI TDRM.
Logistic regression was used to assess factors associated with TDRM carriage and carriage of specific TDRMs by drug class. Factors included in this analysis were birthplace (FSU, SSA, Israel or other), HIV-1 subtype (A, C or non A/C), viral load, age at diagnosis and year of diagnosis (supplemental Table S1). Significant association between recent diagnosis and NNRTI TDRM as found by both univariate (OR: 1.23, 1.05-1.45 of 95% CI, p=0.01) and multivariate analysis (OR: 1.23, 1.03-1.43 of 95% CI, p=0.020). Other associations could not be found.
Table 3 lists the type of TDRMs identified in the study cohort according to drug class. The polymorphic RT-E138 and the accessory mutation A62 sites, were also included due to their clinical relevance and high prevalence. , . A62V which was the most prominent NRTI mutation (5.6%, 19/337), was significantly more common in HIV-1 subtype A- as compared to HIV-1 subtype C-infected women (13%, 17/130 versus 1.4%, 2/141, p<0.001). E138 was the most frequently identified mutated NNRTI position (5.6%, 19/337), detected in 8.5% (n=11), 4.3% (n=6), 3% (n=1) and 4.3% (n=1) of subtype A, C, B and G/AG carriers, respectively. The NNRTI K103N/S mutation was identified in 4.2% (14/337) of women, and was significantly more prominent in those carrying HIV-1 subtype B compared to those carrying subtype C (11.8%, 4/34 versus 2.1%, 3/141, p=0.010). The most prominent PI mutation was M46I, identified in 1.5% (5/337) of patients.
Table 3: Prevalence of most frequently detected TDRM (including NRTI A62 and NNRTI E138) in women, 2010-2018
Drug Class
|
DRM
|
HIV-1 Subtype
|
p value
|
p value
|
p value
|
All
|
A
|
C
|
B
|
G/AG
|
Other
|
A vs. C
|
B vs. C
|
A vs. B
|
N=337
|
N=130
|
N=141
|
N=34
|
N=23
|
N=9
|
|
|
|
PI, n (%)
|
D30N
|
1 (0.3)
|
|
1 (0.7)
|
|
|
|
|
|
|
M46I
|
5 (1.5)
|
1 (0.8)
|
4 (2.8)
|
|
|
|
0.222
|
|
|
V82MS
|
1 (0.3)
|
|
|
|
1 (1.5)
|
|
|
|
|
NRTI, n (%)
|
M41L
|
1 (0.3)
|
|
1 (0.7)
|
|
|
|
|
|
|
A62V
|
19 (5.6)
|
17 (13)
|
2 (1.4)
|
|
|
|
<0.001
|
|
|
D67EGN
|
5 (1.5)
|
3 (2.3)
|
2 (1.4)
|
|
|
|
0.582
|
|
|
K70R
|
3 (0.9)
|
2 (1.5)
|
1 (0.7)
|
|
|
|
0.526
|
|
|
M184V
|
5 (1.5)
|
3 (2.3)
|
2 (1.4)
|
|
|
|
0.582
|
|
|
T215EIS
|
4 (1.2)
|
1 (0.8)
|
3 (2.1)
|
|
|
|
0.376
|
|
|
K219Q
|
1 (0.3)
|
1 (0.8)
|
|
|
|
|
|
|
|
NNRTI, n (%)
|
K101E
|
1 (0.3)
|
|
1 (0.7)
|
|
|
|
|
|
|
K103NS
|
14 (4.2)
|
5 (3.8)
|
3 (2.1)
|
4 (11.8)
|
2 (8.7)
|
|
0.407
|
0.010
|
0.068
|
V106M
|
1 (0.3)
|
|
1 (0.7)
|
|
|
|
|
|
|
E138AGKQ
|
19 (5.6)
|
11 (8.5)
|
6 (4.3)
|
1 (3)
|
1 (4.3)
|
|
0.156
|
0.731
|
0.276
|
Y181C
|
5 (1.5)
|
1 (0.8)
|
4 (2.8)
|
|
|
|
0.222
|
|
|
Y188L
|
1 (0.3)
|
1 (0.8)
|
|
|
|
|
|
|
|
G190AS
|
6 (1.8)
|
2 (1.5)
|
4 (2.8)
|
|
|
|
0.465
|
|
|
Data are presented as n (%). Differences in proportions were measured using the chi-squared test. Empty cells, n=zero.