Malian Breast Cancer Study
We evaluated the association between PIN3 16-bp duplication polymorphism of TP53 gene and the occurrence of breast cancer risk Malian women. The socio-demographic, clinical, and pathological characteristics of the patients are shown in Table 1. The mean age of cases and controls was 43.72 ± 3.14 years and 43.90 ± 2.92 respectively. The distribution of breast cancer in patients age ≤ 40 years (29/60) was similar to patients age > 40 years (31/60). Breast cancers in this cohort were largely within the left breast (37/60). Most women were not using a contraceptive (45/60), not menopausal (29/60), premenopausal (11/60), post-menopausal (20/60). Multiparity was reported (45/60), had or were breastfeed (53/60), did not have a family history of breast cancer (52/60) or any prior history of benign breast disease (54/60), were not obese (41/60) and were not a smoker (53/60). Invasive ductal carcinoma forms (56/60) were more prevalent than any other histological form of breast cancer (Glycogen-rich clear cell carcinoma, lobular carcinoma in situ, Moderately differentiated adenocarcinoma and infiltrating adenocarcinoma). Patients with PIN3 16 bp duplication (A2A2) of TP53 gene were significantly more likely to have an invasive ductal carcinoma form, T3 stage tumor size, node involvement (N0 and N1), and M0 metastasis status compared to patients with the A1A1 or A1A2 genotype. With the exception of histological types (p = 0.04), a negative correlation was observed between the 16-bp polymorphism duplication and the other characteristics such as age, tumor location, contraceptive history, menopausal status, parity, maternal breastfeeding, family history of breast cancer, family history of other cancer, benign breast involvement, obesity, smoking, tumor size, lymph node invasion and metastasis status
Table 1
Distribution of the PIN3 16-bp duplication polymorphism of TP53 gene according to the clinicopathological characteristics in Malian breast cancer.
| | PIN3 16-bp duplication | | |
Clinical Parameter | N | A1A1 | A1A2 | A2A2 | X2 | P |
Age of diagnostic | 60 | | | | 2.41* | 0.12 |
≤ 40 years of age | 29 | 11 (37.9) | 12 (41.4) | 6 (20.7) | | |
> 40 years of age | 31 | 16 (51.6) | 13 (41.4) | 2 (6.5) | | |
Localization | | | | | 1.98 | 0.74 |
Right breast | 19 | 7 (36.8) | 9 (47.4) | 3 (15.8) | | |
Left breast | 37 | 19 (51.4) | 14 (37.8) | 4 (10.8) | | |
Bilateral | 4 | 1 (25.0) | 2 (50.0) | 1 (25.0) | | |
Use of contraceptives | | | | 0.56* | 0.45 |
No | 45 | 18 (40.0) | 25 (55,6) | 2 (4.4) | | |
Yes | 15 | 9 (60.0) | - | 6 (40.) | | |
Menopausal status | | | | | 3.15 | 0.53 |
Pre-menopausal | 11 | 6 (54.5) | 4 (36.4) | 1 (9.1) | | |
Post-menopausal | 20 | 10 (50.0) | 9 (45.0) | 1 (5.0) | | |
Fertile women | 29 | 11 (37.9) | 12 (41.4) | 6 (20.7) | | |
Parity | | | | | 7.33 | 0.12 |
Nulliparity | 6 | - | 5 (83.3) | 1 (16.7) | | |
Primiparity | 9 | 3 (33.3) | 4 (44.4) | 2 (22.2) | | |
Multiparity | 45 | 24 (53.3) | 16 (35.6) | 5 (11.1) | | |
Breastfeeding | | | | | 0.50* | 0.48 |
Yes | 53 | 26 (49.1) | 19 (35.8) | 8 (15.1) | | |
No | 7 | 1 (14.3) | 6 (85.7) | - | | |
Family history of BC | | | | | 0.64* | 0.42 |
Yes | 8 | 4 (50.0) | 4 (50.0) | - | | |
No | 52 | 23 (44.2) | 21 (40.4) | 8 (15.4) | | |
History of benign breast disease | | | | 1.69* | 0.19 |
Yes | 6 | 4 (66.7) | 2 (33.3) | - | | |
No | 54 | 23 (42.6) | 23 (42.6) | 8 (14.8) | | |
Obesity | | | | | 0.00* | 0.99 |
Yes | 19 | 8 (42.1) | 9 (47.4) | 2 (10.5) | | |
No | 41 | 19 (46.3) | 16 (39.0) | 6 (14.6) | | |
Smoking | | | | | 0.20* | 0.65 |
Passive smoking | 7 | 3 (42.9) | 4 (57.1) | - | | |
No | 53 | 24 (45.3) | 21 (39.6) | 8 (15.1) | | |
Histological type | | | | | 4.14* | 0.04 |
Invasive ductal carcinoma | 56 | 23 (41.1) | 25 (44.6) | 8 (14.3) | | |
Other | 4 | 4 (100.0) | - | - | | |
Tumor size | | | | | 5.63 | 0.46 |
T1 | 1 | - | 1 (100.0) | - | | |
T2 | 10 | 5 (50.0) | 5 (50.0) | - | | |
T3 | 41 | 18 (43.9) | 15 (36.6) | 8 (19.5) | | |
T4 | 8 | 4 (50.0) | 4 (50.0) | - | | |
Nodal involvement | | | | | | |
N0 | 36 | 16 (44.4) | 16 (44.4) | 4 (11.1) | 6.05 | 0.41 |
N1 | 16 | 5 (31.3) | 7 (43.8) | 4 (25.0) | | |
N2 | 7 | 5 (71.4) | 2 (28.6) | - | | |
N3 | 1 | 1 (100.0) | - | - | | |
Metastasis | | | | | 0.91* | 0.34 |
M0 | 55 | 24 (43.6) | 23 (41.8) | 8 (14.5) | | |
M1 | 5 | 3 (60.0) | 2 (40.0) | - | | |
X2: Chi-squared test; P = p-value; *: Chi-squared test for trend; N: Number; BC: Breast cancer, A1A1: Wild-type; A1A2: heterozygous; A2A2: mutated homozygous or 16-bp duplication. Other: Glycogen-rich clear cell carcinoma, lobular carcinoma in situ, Moderately differentiated adenocarcinoma and infiltrating adenocarcinoma |
PIN3 16-bp Duplication Polymorphism of TP53 Gene and Breast Cancer Risk
Table 2 shows the distribution of PIN3 16-bp duplication polymorphism of the TP53 gene in the cases according to the genetic models. The genotypic distribution PIN3 16-bp polymorphism is aligned with the Hardy-Weinberg equilibrium in cases (X2 = 0.33, p = 0.57) and controls (X2 = 2.76, p = 0.10). Genotypic frequencies in cases and controls were respectively 45.0% A1A1, 41.7% A1A2, 13.3% A2A2 and 65.0% A1A1, 26.7% A1A2, 8.3% A2A2. In addition, allelic frequencies were 65.8% A1 and 34.2% A2 in cases vs. 78.3% A1 and 21.7% A2 in controls (Table 2). A1A2 heterozygote has been associated with a high risk of breast cancer with an OR of 2.25 (1.01–5.01) and p = 0.04. For all of the genetic combinations, except for the recessive model (A2A2 vs. A1A1 + A1A2: OR = 1.69, CI 95% = 0.52–5.50, p = 0.38), PIN 3 16 bp duplication polymorphism was associated with an increased risk of breast cancer, particularly in the dominant (A1A2 + A2A2 vs. A1A1: OR = 2.26, CI 95% = 1.08–4.73, p = 0.02) and additive (A2 vs A1 : OR = 1.87, CI 95% = 1.05–3.33, p = 0.03) models (Table 2).
Table 2
Distribution of PIN3 16-bp duplication polymorphism of the TP53 gene in Malian breast cancer according genetic models.
Genotype/Allele | Cases | Controls | | |
N = 60 | N = 60 | OR (95% CI) | P |
A1A1 | 27 (45.0) | 39 (65.0) | Reference | |
A1A2 | 25 (41.7) | 16 (26.7 | 2.25 (1.01–5.01) | 0.04* |
A2A2 | 8 (13.3) | 5 (8.3) | 2.31 (0.68–7.83) | 0.17 |
A2A2 + A1A2 | 33 (55.0) | 21 (35.0) | 2.26 (1.08–4.73) | 0.02* |
A1A1 + A1A2 | 52 (86.7) | 55 (91.7) | Reference | |
A2A2 | 8 (13.3) | 5 (8.3) | 1.69 (0.52–5.50) | 0.38 |
A1 | 79 (65.8) | 94 (78.3) | Reference | |
A2 | 41 (34.2) | 26 (21.7) | 1.87 (1.05–3.33) | 0.03* |
N = Number; CI: confidence Interval; P: p-value; A2A2 + A1A2 vs. A1A1: Dominant model, A2A2 vs. A1A1 + A1A2: Recessive model; A2 vs. A1: Additive model; *: Significance. |
Meta-analysis Study
Characteristics of Included Studies
A total of 19 articles reporting on case-control studies that investigated PIN3 16 duplication polymorphism and breast cancer risk and satisfied the inclusion criteria (Table 3) were included in the meta-analysis. Thirty-seven studies with no PIN3 16 bp duplication of TP53 gene, 6 studies with deviation from HWE and 2 studies [27, 28] with influence on the values of the pooled OR and p-value, were excluded.
Table 3
Characteristics of PIN3 16-bp polymorphism genotype distributions of TP53 gene in studies included in this meta-analysis.
| | Cases | Controls | |
Authors | Population | N | A1A1 | A1A2 | A2A2 | N | A1A1 | A1A2 | A2A2 | HWE |
Present study | Mali | 60 | 27 | 25 | 8 | 60 | 39 | 16 | 5 | 0.10 |
Akkiprik et al 2009 [19] | Turkey | 97 | 59 | 35 | 3 | 107 | 61 | 43 | 3 | 0.15 |
Buyru et al 2007 [29] | Turkey | 115 | 83 | 28 | 4 | 63 | 47 | 15 | 1 | 0.87 |
Cherdyntseva et al 2012 [30] | Russia | 296 | 227 | 68 | 1 | 196 | 145 | 50 | 1 | 0.13 |
Costa et al 2008 [18] | Portugal | 191 | 122 | 56 | 13 | 216 | 147 | 65 | 4 | 0.29 |
De Vecchi et al 2008 [31] | Italy | 350 | 233 | 103 | 14 | 352 | 256 | 87 | 9 | 0.62 |
Gaudet et al 2007 [32] | USA (M) | 578 | 404 | 157 | 17 | 390 | 272 | 108 | 10 | 0.85 |
Gohari-Lasaki et al 2015 [24] | Iran | 100 | 53 | 38 | 9 | 100 | 60 | 37 | 3 | 0.34 |
Guleria et al 2012 [33] | India | 80 | 43 | 30 | 7 | 80 | 53 | 25 | 2 | 0.64 |
Hao et al 2018 [34] | Chine | 254 | 230 | 24 | 0 | 252 | 227 | 25 | 0 | 0.41 |
Hrstka et al 2009 [35] | Island | 117 | 81 | 32 | 4 | 108 | 81 | 24 | 3 | 0.46 |
Morten et al 2019 [21] | Australia | 1304 | 986 | 289 | 29 | 436 | 325 | 104 | 7 | 0.67 |
Pouladi et al 2014 [36] | Iran | 221 | 135 | 69 | 17 | 170 | 107 | 51 | 12 | 0.10 |
Sharma et al 2014 [7] | India | 200 | 134 | 52 | 14 | 200 | 137 | 55 | 8 | 0.41 |
Suspitsin et al 2003 [37] | Russia | 529 | 408 | 108 | 13 | 249 | 187 | 56 | 6 | 0.47 |
Trifa et al 2010 [38] | Tunisia | 159 | 98 | 56 | 5 | 132 | 86 | 41 | 5 | 0.97 |
Vymetalkova et al 2015 [39] | Czech | 705 | 474 | 164 | 24 | 611 | 421 | 172 | 18 | 0.93 |
Wang-Gohrke et al 2002 [17] | Germany | 563 | 370 | 173 | 20 | 549 | 391 | 145 | 13 | 0.92 |
Weston et al 1997 [40] | USA (M) | 99 | 60 | 36 | 3 | 185 | 127 | 54 | 4 | 0.52 |
M : Mixed ; N = Number |
Quantitative Analysis
This meta-analysis showed a significant association between PIN3 16-bp duplicate polymorphism of TP53 gene and breast cancer in recessive (Fixed effect model (FEM): OR = 1.46, 95% CI = 1.15–1.85; p = 0.002) and additive (FEM: OR = 1.11, 95% CI = 1.02–1.19; p = 0.01) models, but not in a dominant (FEM: OR = 1.07, 95% CI = 0.98–1.20; p = 0.15) model. Figures 2, 3, and 4, respectively, show the forest plots of OR for breast cancer in the dominant, recessive and additive models of PIN3 16-bp duplication polymorphism of the TP53 gene, after excluding articles with HWE deviation of controls.
Sensitivity Analysis
The stability of the results was assessed by a sensitivity analysis. After withdrawal of studies that were not in the Hardy-Weinberg equilibrium and the exclusion of studies influencing the values of the pooled OR [27] and p-value [28], with the exception of the dominant model (Fig. 2), a significant association was observed between the PIN3 16-bp and breast cancer in the recessive (Fig. 3) and additive (Fig. 4) models. Similarly, the one by one elimination of eligible studies does not influence the pooled effect OR values in the different genetic models.
Sources of Heterogeneity
After non-inclusion of articles with HWE-deviation in controls, a lack of heterogeneity was found in the dominant (I2 = 19%, P = 0.23), recessive (I2 = 0%, P = 0.94) and additive (I2 = 11%, P = 0.32) models between PIN3 16-bp duplicate polymorphism and breast cancer risk (Figs. 1, 2, and 3). However, this heterogeneity test was not statistically significant in the different genetic models.
Publication Bias
A funnel plot was used to evaluate publication bias. After non-inclusion of articles witih HWE deviation in controls and sensitivity analysis, no publication bias was observed in the recessive and additive models. However, a slight asymmetry was detected in the dominant model (Fig. 5).