With the widespread application of screening mammography, more ductal carcinoma in situ (DCIS) has been detected over the past few decades. Currently, among all the newly-diagnosed breast cancer, one fifth was presented as DCIS (1, 2). Although DCIS was considered a rather indolent lesion itself, approximately 25% to 50% of them will progress into invasive ductal carcinoma (IDC) eventually. So far, the treatment backbone for DCIS is still surgery, in a similar manner as IDC tumor.

Elderly patients usually were presented with more comorbidities, and have relatively shorter life expectancies (3). Normally, a trend of treatment de-escalation exists among elderly breast cancer patients. Elderly patients with DCIS experience lower local recurrence rate than younger patients (4–6), therefore debate remains about how to select the optimal treatment for them. Some suggests that active surveillance may be safe for elderly patients with rather low risk DCIS, in order to avoid overtreatment and reduce morbidity caused by surgery. However, others argue that elderly patients had longer life expectancies now and should be treated with same standard as younger patients (7). Notably, elderly patients themselves are heterogeneous, with or without co-existing illness, and different kinds of illness all render them into different physical condition, resulting in different tolerance of local treatment. Currently, little is known regarding the clinical and pathological factors that contribute to treatment decisions in elderly DCIS patients.

Based on above issue, this study aims to evaluate the current local treatment patterns of elderly patients with DCIS. Also, we aim to explore the factors that influence the choice of local treatment and their associations with prognosis for elderly DCIS patients.

The relatively indolent nature and good prognosis of DCIS raises concerns on its over-diagnosis and overtreatment, especially among elderly patients (9, 10). In this study, we found that age rather than comorbidity status significantly influence the choice of local treatments. Elderly DCIS patients appear to receive less aggressive surgery type and less adjuvant radiotherapy after BCS, but without impaired disease outcome.

Currently it's acknowledged that DCIS is a precursor lesion to most, if not all, invasive breast cancer (11). However, this progression is usually unpredictable and a considerable percentage of DCIS lesions will never become invasive (2). Compared with younger DCIS patients, the recurrence rate of DCIS in older women is lower. Considering limited life expectancy in elderly patients, less aggressive treatments are usually recommended (12, 13). However, the appropriate local treatment for elderly patients with DCIS is still controversial. Debate remains about the feasibility to choose active monitoring in substitution for surgery, to omit radiotherapy after BCS, or to spare axillary evaluation during mastectomy (14). Moreover, evidence is still scarce regarding factors influencing the choice of different local treatment.

Surgery is still regarded as the primary treatment for DCIS tumors. Although DCIS patients are eligible for either mastectomy or breast conserving surgery with equivalent safety and survival benefit (15), more than half of cases in our cohort underwent mastectomy, which is consistent with previous reports (16). Bleicher et al. found that older women with DCIS chose mastectomy over breast-conserving treatment if they have larger tumor size, lower education level, or consulted greater number of surgeons, while age and comorbidities did not predict choosing mastectomy (3). According to our results, patients older than 80 years were more likely to receive BCS than mastectomy. Moreover, patients with tumor size larger than 1.5 cm or primarily presented with mass symptom were less likely to receive BCS. This was not unexpected as for larger tumors, as BCS may be difficult to achieve clear resection margin. Among well-established risk factors for local recurrence in DCIS including histologic subtype, nuclear grade, and age, etc., margin status was described as the most important one (17, 18). Although tumor size was not identified as the predicting factor for LRR in DCIS according to NSABP B-17 or EORTC trials (19, 20), it is anticipated that larger tumor size and non-mass lesion might indicate an extensive lesion requiring mastectomy to ensure clear margin status. In this study, more than half (71%) patients received preoperative breast MRI evaluation, which might identify non-mass enhancement beyond target lesion found by mammography or ultrasound. Application of preoperative MRI were probably associated with decreased breast-conserving rate in this study. In addition, co-existing cardiovascular diseases would raise concern when considering radiotherapy after BCS. Patient's preference is an important factor that determine surgery type in China. Surgery type, BCS or mastectomy, was usually discussed by patients and her family members. Elderly patients care less about cosmetics but more about side effects and economics of radiotherapy. Therefore, mastectomy could be an option (21, 22).

As recommended, mastectomy is routinely accompanied by axillary evaluation for DCIS cases because subsequent sentinel lymph node biopsy would be difficult to perform if an invasive disease was found on postoperative pathological specimen (23). Especially in patients diagnosed with core needle biopsy since limited sample may lead to pathological underestimation (24). Consistently, in our study, compared with patients receiving excisional biopsy, patients receiving core needle biopsy prior to surgery were more likely to underwent axillary evaluation. We also observed that the percentage of patients receiving ALN surgery were significantly higher in mastectomy subgroup than BCS subgroup.

Less axillary evaluation was performed in elderly patients according to our results, especially for those older than 80 years. Furthermore, our study demonstrated that receiving ALN surgery or not have no impact on local recurrence. DCIS patients usually have no clinically detected lymph node. Although the final pathological diagnosis might be upgraded to invasive cancer, axillary lymph node metastasis and regional recurrence is still scarce for DCIS patients (19, 25). We admitted that the proportion of patients who received ALN surgery is relatively high. However, the real-world clinical practice in China is somewhat difficult to follow the treatment standard of DCIS, which did not recommend routine axillary evaluation for DCIS patients, especially for those received BCS. Chinese patients usually refuse a secondary ALN surgery if invasive disease is detected pathologically after primary surgery. Therefore, most patients demand ALN evaluation at the same time when they received breast surgery. Some radical patients request for a total mastectomy and axillary lymph node dissection even if no invasive breast cancer was found in preoperative biopsy. Though there exist the worries on a second surgery or locoregional recurrence from patients, the rate of upstaging from DCIS to invasive disease has been reported less than 20%. Most upstaged disease were Stage IA invasive ductal carcinoma, which have low risk of nodal metastasis (26). In recent years, we took great effort in patient education and found that the proportion of ALN surgery decreased in DCIS patients. Therefore, it is reasonable to presume ALN surgery could be omitted when performing BCS for elderly DCIS patients.

The benefit brought by postoperative radiotherapy in DCIS patients must be carefully weighed against the accompanying complications (27). With the development of modern radiation techniques, radiotherapy has been proved to be safe and has minimal impact on quality of life, and leading to limited cardiovascular mortality for the elderly (28–30). However, worries on deterioration of their comorbidities and inconvenient daily hospital visits for radiotherapy still trouble specialists and patients. According to our results, more than half of all patients were omitted of postoperative radiotherapy after BCS, and for patients older than 80 years, up to 84.6% were precluded with adjuvant radiotherapy. Age was the only factor related to radiotherapy after BCS. Consistently, Smith et al. had also observed that the proportion of receiving adjuvant radiotherapy decreased while patient age increased (31). Their study reported an omission of radiotherapy after BCS in 51.0% of all patients and 36.8%, 49.9% and 70.8% in patients aging 66–69, 70–79 and ≥80, respectively (P < 0.001).

A number of randomized clinical trials have already demonstrated that adding radiotherapy after surgery for DCIS patients of all ages could improve local control rate, which could reduce IBTR by approximately half (19, 32–36). However, the survival benefit brought by radiotherapy in elderly patients, especially patients over 80 years remained controversial. Smith et al. (31) found that radiotherapy after BCS contributed to a significant reduction in LRR in a group of DCIS patients over 65 years old. According to age subgroups, they found that healthy women of 66–79 years old were twice as likely to benefit from radiotherapy than patients ≥85 years who have moderate to severe comorbidity, leaving the benefit of radiotherapy for patients with rather old age no less debatable (31). Also, an EBCTCG meta-analysis had showed that radiotherapy resulted in a greater reduction in LRR for DCIS patients older than 50 years when compared with younger women, while no further study was conducted among patients with age over 65 or more (36).

On the contrary of supporting adjuvant radiotherapy for elderly patients, a study by Ho et al.(37) reported no LRR difference between patients receiving radiotherapy or not in women older than 60 years. Likewise, our study also found that elderly patients with different local treatment modalities (mastectomy, BCS plus radiotherapy, or BCS alone) shared similar LRR. Moreover, none of the patients ≥80 years in our study experienced LRR during follow-up. According to our inclusion criteria and clinicopathological characteristics, the high percentage of negative margin status and low Ki-67 index of enrolled patients probably reduced the potential benefit from radiotherapy (38). The role of adjuvant radiotherapy for elderly DCIS patients warrants further investigation since available evidence is limited.

DCIS is a group of disease with heterogeneous natural course and prognosis. A lot of effort has been made in risk stratification in order to identify a group of DCIS patients with good prognosis, in whom surgical excision alone or even observation could be enough to achieve a satisfactory local control. According to available prognostic factors for DCIS, older age could predict decreased risk of recurrence. Prognostic scores or multigene assays could also be used to evaluated the local control benefit offered by radiotherapy after BCS in DCIS patients (22, 39–42). In future perspective, local treatment strategies may be tailored according to recurrence stratification model in order to balance benefit and risk.

Our study has several limitations. Firstly, the number of patients included in our study was limited. And our follow-up time is relatively short, given the long natural history of DCIS. Therefore, the small number of outcome events may not provide sufficient statistical power to detect the benefit conferred by treatment. Secondly, this is a single institution retrospective study. Large-scaled prospective studies are warranted to validate our results.

In summary, our study presents the current approach of local treatment in elderly DCIS patients. Age is related with the choice of breast surgery, ALN surgery, and postoperative radiotherapy. DCIS patients with age ≥80 years old receive less aggressive local treatments but have no impaired disease outcome.