Faculté de Médecine de Sousse, Farhat Hached University Hospital, Department of Medical Oncology, Association de Recherche et d'Information sur le Cancer du Centre Tunisien (ARIC), Université de Sousse, 4000, Sousse, Tunisia.
K9 Dog Security and Training Center, Sousse, Tunisia.
BMC Cancer. 2024 Jan 30;24(1):151. doi: 10.1186/s12885-023-11599-3.
Breast cancer in Tunisia is often diagnosed at a late stage with long delay in time to consultation and to diagnosis.The aim of this study is to estimate the sensitivity and specificity of the transcutaneous breast cancer detection by canine olfactionin Tunisian women and to identify the potential confounding factors.
This is a diagnostic case control study that took place from October 2021 to November 2022 in the Department of Medical Oncology at the University Hospital Farhat Hached of Sousse and in the security and training dog center located in Sousse (K9 Dog Center Security & Training). A two-year-old male Belgian Malinois was trained to detect breast cancer on skin secretion samples in compresses that had been worn overnight by women on their breast and then a double-blind testing was performed. There was no contact between women and the dog. From the mentioned responses of the dog, four parameters were calculated: sensitivity, specificity, Positive Predictive Value (PPV) and Negative Predictive Value (NPV).
Two hundred women were included in this trial: 100 breast cancer (BC) patients recruited from Farhat Hached University Hospital of Sousse and 100 healthy volunteers (HV).The calculated sensitivity was 84% (95% CI 78-89%) and the calculated specificity was 81% (95% CI 75-86%). The calculated predictive values were: PPV = 83,51% (95% CI 78,37-88,65%) and NPV = 81,55% (95% CI 76.17-86.93%). In the multivariate study, only four confounding factors of test's sensitivity were retained: age (OR = 1.210 [95% CI = 1.085-1.349]; p = 0.001), history of diabetes(OR = 0.017 [95% CI = 0.001-0.228]; p = 0.002), sampling at hospital (OR = 0.010 [95% CI = 0.003-0.464]; p = 0.010) and testing during chemotherapy courses (OR = 0.034 [95% CI = 0.003-0.404]; p = 0.007).For test's specificity, we retained the three following confounding factors: age (OR = 1,104 [95% CI = 1.021-1.195]; p = 0.014), history of benign mastopathy (OR = 0.243 [95% CI = 0.074-0.805]; p = 0.021)and history of arterial hypertension (OR = 0.194 [95% CI = 0.053-0.707]; p = 0.013).
This is a pilot study that opens new avenues in developing a reliable cancer diagnostic tool that integrates the dog's olfactory ability to detect breast cancer using a transcutaneous sampling method. It could be a pre-test to select patients who are eligible to a screening mammogram, especially in low-income countries where there is no national mammography screening program. PACTR.
PACTR202201864472288, registration date 11/01/2022.
突尼斯的乳腺癌通常在晚期诊断，咨询和诊断的时间间隔较长。本研究的目的是估计犬类嗅觉经皮乳腺癌检测的敏感性和特异性，并确定潜在的混杂因素。
这是一项诊断性病例对照研究，于 2021 年 10 月至 2022 年 11 月在苏塞的法哈特·哈切德大学医院和苏塞的安全和培训犬中心（K9 犬中心安全与培训）进行。一只两岁的雄性比利时马里努阿犬经过训练，能够检测女性在乳房上佩戴过夜的压缩包中的皮肤分泌物样本中的乳腺癌，然后进行双盲测试。女性与狗没有任何接触。根据狗的反应，计算了四个参数：敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV）。
共有 200 名女性参与了该试验：100 名乳腺癌（BC）患者来自苏塞法哈特·哈切德大学医院，100 名健康志愿者（HV）。计算的敏感性为 84%（95%CI 78-89%），特异性为 81%（95%CI 75-86%）。计算的预测值为：PPV=83.51%（95%CI 78.37-88.65%）和 NPV=81.55%（95%CI 76.17-86.93%）。在多变量研究中，仅保留了四个测试敏感性的混杂因素：年龄（OR=1.210 [95%CI 1.085-1.349]；p=0.001）、糖尿病史（OR=0.017 [95%CI 0.001-0.228]；p=0.002）、在医院采样（OR=0.010 [95%CI 0.003-0.464]；p=0.010）和在化疗期间测试（OR=0.034 [95%CI 0.003-0.404]；p=0.007）。对于测试的特异性，我们保留了三个混杂因素：年龄（OR=1.104 [95%CI 1.021-1.195]；p=0.014）、良性乳腺疾病史（OR=0.243 [95%CI 0.074-0.805]；p=0.021）和高血压病史（OR=0.194 [95%CI 0.053-0.707]；p=0.013）。
这是一项初步研究，为开发一种可靠的癌症诊断工具开辟了新途径，该工具集成了狗的嗅觉能力，通过经皮取样方法检测乳腺癌。它可以作为一种预测试，以选择有资格进行乳房 X 线筛查的患者，特别是在没有国家乳房 X 线筛查计划的低收入国家。PACTOR.ORG 标识符：PACTOR202201864472288，注册日期 2022 年 1 月 11 日。
