【原】年龄、亚型、乳腺癌女性死亡风险

　　年龄和肿瘤亚型是乳腺癌相关生存的预后因素，但是目前尚不清楚哪个最重要。由于乳腺癌患者的数量不断增加，故有必要根据全部临床患者人群样本，明确所有年龄分组和临床亚型的乳腺癌所致死亡风险。

　　2018年12月3日，国际抗癌联盟《国际癌症杂志》在线发表挪威癌症登记中心、奥斯陆大学、瑞典卡罗林学院、英国伦敦帝国学院的大数据研究报告，通过挪威全国年轻与年老乳腺癌女性7年随访，对不同临床亚型的乳腺癌相关生存进行了分析。

　　该研究通过挪威癌症登记数据库，确定2005～2015年被诊断为乳腺癌年龄20～89岁女性共计2万1384例。根据雌激素受体（ER）、孕激素受体（PR）和人表皮生长因子受体2（HER2）状态，将乳腺癌亚型定义为：

• 管腔A型（ER+ PR+ HER2-）

• 管腔B1型（ER+ PR- HER2-）

• 管腔B2型（ER+ PR± HER2+）

• HER2阳性（ER- PR- HER2+）

• 三阴性（ER- PR- HER2-）

　　通过多因素比例风险回归模型分析，按不同的年龄和亚型，推算7年乳腺癌所致死亡风险比，同时按不同的诊断年份、分级、TNM分期和治疗方案等影响因素进行校正。

　　结果发现，HER2阳性和三阴性乳腺癌多见于年轻女性，而管腔A型乳腺癌多见于70～89岁的年老女性。

　　按不同的诊断年份进行校正后，与挪威乳腺癌筛查年龄50～59岁女性相比：

• 20～39岁女性的乳腺癌所致死亡风险高2.26倍（95%置信区间：1.81～2.82）

• 40～49岁女性的乳腺癌所致死亡风险相似

• 60～69岁女性的乳腺癌所致死亡风险相似

• 70～79岁女性的乳腺癌所致死亡风险高2.25倍（95%置信区间：1.87～2.71）

• 80～89岁女性的乳腺癌所致死亡风险高5.19倍（95%置信区间：4.21～6.41）

　　按不同的诊断年份、年龄、亚型、分级、分期、手术方案进行校正后，20～39岁女性的乳腺癌所致死亡风险并不显著，70～89岁女性的乳腺癌所致死亡风险仍然较高。

　　因此，该研究结果表明，年轻与管腔A型乳腺癌所致死亡风险增加相关，而年老与所有亚型乳腺癌所致死亡风险增加相关。年龄和亚型是强而独立的预后因素。按不同亚型进行校正后，年老女性的乳腺癌所致死亡风险始终较高，原因可能在于年老女性的合并症较多。肿瘤本身因素（亚型、分级和分期）在很大程度上可以解释年轻女性的乳腺癌所致死亡风险较高，故有必要开展进一步研究，解答为何年轻女性的管腔A型乳腺癌所致死亡风险较高。

Int J Cancer. 2018 Dec 3. [Epub ahead of print]

Breast cancer-specific survival by clinical subtype after 7 years follow-up of young and elderly women in a nationwide cohort.

Johansson ALV, Trewin CB, Hjerkind KV, Ellingjord-Dale M, Johannesen TB, Ursin G.

Cancer Registry of Norway, Oslo, Norway; Karolinska Institutet, Stockholm, Sweden; Imperial College London, London, UK; University of Oslo, Norway.

What's new? Given the increasing numbers of patients with breast cancer, it is important to provide solid estimates of the breast cancer-specific mortality across all age groups and clinical subtypes based on population-based samples representative of the full spectrum of patients in the clinic. Using national cancer registry data, here the authors found that young women (<40) had a higher breast cancer-specific mortality compared to screening-aged women (50-69), in particular among luminal A-like tumors, while older women (70-89) had a higher breast cancer-specific mortality within all subtypes of breast cancer. Comorbidity is a potential explanation for the remaining survival deficit in elderly women.

Age and tumor subtype are prognostic factors for breast cancer survival, but it is unclear which matters the most. We used population-based data to address this question. We identified 21,384 women diagnosed with breast cancer at ages 20-89 between 2005 and 2015 in the Cancer Registry of Norway. Subtype was defined using estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor 2 (HER2) status as luminal A-like (ER+PR+HER2-), luminal B-like HER2-negative (ER+PR-HER2-), luminal B-like HER2-positive (ER+PR+/-HER2+), HER2-positive (ER-PR-HER2+) and triple-negative (TNBC) (ER-PR-HER2-). Cox regression estimated hazard ratios (HR) for breast cancer-specific 7-year survival by age and subtype, while adjusting for year, grade, TNM stage and treatment. Young women more often had HER2-positive and TNBC tumors, while elderly women (70-89) more often had luminal A-like tumors. Compared to age 50-59, young women had doubled breast cancer-specific mortality rate (HR = 2.26, 95% CI 1.81-2.82), while elderly had two to five times higher mortality rate (70-79: HR = 2.25, 1.87-2.71; 80-89: HR = 5.19, 4.21-6.41). After adjustments, the association was non-significant among young women but remained high among elderly. Young age was associated with increased breast cancer-specific mortality among luminal A-like subtype, while old age was associated with increased mortality in all subtypes. Age and subtype were strong independent prognostic factors. The elderly always did worse, also after adjustment for subtype. Tumor-associated factors (subtype, grade and stage) largely explained the higher breast cancer-specific mortality among young. Future studies should address why luminal A-like subtype is associated with a higher mortality rate in young women.

KEYWORDS: IHC markers; age; breast cancer; clinical subtype; survival

PMID: 30367449

DOI: 10.1002/ijc.31950