Stage 0 Breast Cancer (DCIS, LCIS)

What every physician needs to know:

Stage 0 breast cancer includes non-invasive breast cancer; ductal carcinoma in situ (DCIS), and lobular carcinoma in situ (LCIS). Physicians should understand the following:

• Incidence and clinical significance of DCIS and LCIS.

• Pathologic findings of both DCIS and LCIS.

  ---

  Continue Reading

• Differences in management of DCIS versus LCIS.

Incidence of in situ cancer

In 2012, 63,300 new cases of in situ cancer will be diagnosed in the United States compared to the 229,060 cases of invasive cancer. Often when breast cancer statistics are quoted, these in situ cases are not included, despite the fact that they represent an increasing proportion of breast cancer cases. In the early 1980’s, in situ cancer accounted for 3-4% of all breast cancers compared to 22% currently.

Most in situ cancers are DCIS. DCIS accounted for 83% of in situ cancers diagnosed between 2004-2008. The incidence of DCIS has been increasing over time. The DCIS rate increased about 2% per year between 1973 and 1982. From 1982 to 1988, the DCIS rate increased dramatically, about 28% per year, at the same time mammography screening was increasing. After this, the DCIS rate increased more slowly at about 6% per year.

Since 1999, incidence rates of in situ breast cancer stabilized among women over 50, but continue to increase in younger women. The stabilization in incidence among women over 50 years of age may correlate with trends in mammography screening, which peaked in 2000.

Are you sure your patient has stage 0 breast cancer? What should you expect to find?

Ductal carcinoma in situ

DCIS is usually asymptomatic and identified first with an abnormal mammogram. In modern times, DCIS will rarely present as a palpable mass. A spontaneous nipple discharge which is bloody, pink tinged, clear or serous in nature may be a presenting symptom.

Lobular carcinoma in situ

LCIS is usually asymptomatic and will be found as an incidental finding on biopsy of the breast for other findings. LCIS is found in approximately 1% of all excisional breast biopsy specimens. About 80% of LCIS occurs in pre-menopausal women (mean age of 49 years).

Beware of other conditions that can mimic stage 0 breast cancer:

Differential diagnosis for pathologic nipple discharge (most often notdue to DCIS):

• papilloma- most common

• duct ectasia

• infection

Which individuals are most at risk for developing stage 0 breast cancer:

Risk factors for DCIS resemble those of invasive breast cancer:

• Women aged over 40.

• Family history of breast cancer.

• BRCA 1 and 2 mutation carriers.

• Early age of menarche.

• Late age of first live childbirth.

• Late menopause.

• Post-menopausal use of estrogen and progesterone hormone replacement therapy.

• Overweight status in post-menopausal women.

• Mammographic density.

LCIS is more likely to occur in younger women and is more common in white women than in black women.

What laboratory and imaging studies should you order to characterize this patient's tumor (i.e., stage, grade, CT/MRI vs PET/CT, cellular and molecular markers, immunophenotyping, etc.) How should you interpret the results and use them to establish prognosis and plan initial therapy?

Ductal carcinoma in situ

The diagnosis of DCIS is usually made on core needle biopsy of a mammographic abnormality. Suspicious findings on mammogram are typically evaluated further with spot compression and magnified mammogram views of the abnormality to better characterize the lesion and to assess the extent of the lesion.

The most common mammographic abnormality is the finding of fine microcalcifications which are clustered and pleomorphic in shape (Figure 1, Figure 2, Figure 3) The calcifications are sometimes oriented in a linear array representing the course of the ductal system. Less common mammographic findings include spiculated densities. A careful physical exam should be done to confirm there are no breast abnormalities or regional adenopathy.

Figure 1.

Figure 2.

Figure 3.

For patients presenting nipple discharge, a ductogram can be performed to identify the area of tumor which can be seen as a filling defect in the duct or a cut off of filling of the duct with contrast, or a narrowing of the duct (Figure 4). The medical history should include a detailed family history of cancer as well as hormonal risk factors.

Figure 4.

Because needle biopsy provides only a sample of the mammographic abnormality, excision of the lesion will provide the final confirmation of the lesion as in situ without evidence of invasive disease. The pathology exam will characterize the histology, nuclear grade, and evidence of necrosis as well as the size of the involved area of DCIS and margin width on resected specimens.

A rigorous pathologic evaluation is critical to rule out invasion. The tumor should be evaluated with immunohistochemical stains to assess for estrogen and progesterone receptors. Adjuvant tamoxifen will be considered for tumors that are hormone receptor positive to reduce risk of local recurrence after breast conserving surgery.

There is interest in identifying markers that could be prognostic for local recurrence after excision alone. One of the markers that has been investigated is Her2neu which has been elucidated in invasive breast cancer and for which there is a targeted drug, trastuzumab. DCIS is positive for Her2Neu expression in approximately 40% of cases. However, at this time, routine examination for Her2neu is not recommended since there is currently no therapeutic intervention made based on these results. There are clinical trials in progress to address the role of trastuzumab in DCIS management.

Lobular carcinoma in situ

The diagnosis of LCIS is usually an incidental finding on excisional biopsy of mammographic lesions or palpable masses. If LCIS is found on a core needle biopsy, excisional biopsy must be performed to confirm that there is no invasive breast cancer in the sampled area.

Bilateral mammograms should be obtained with focused diagnostic views in the area of the abnormality. Ultrasound should be obtained if mass lesions are seen on mammography. A careful physical examination of the breasts and regional nodes should be performed. Medical history should include a detailed family history of cancers as well as hormonal risk factors.

Both LCIS and DCIS are staged as Stage 0, TisN0M0. While tumor size is very important for planning treatment of DCIS, it does not impact the staging determination. While both are Stage 0, DCIS and LCIS are managed quite differently.

Staging

Table I. TNM staging of breast cancer.

What therapies should you initiate immediately i.e., emergently?

Therapy for DCIS and LCIS is not needed emergently. There is always time for appropriate work-up and evaluation to characterize the lesion and ascertain which treatments will be most appropriate. Consultations can be obtained with radiation oncology, plastic surgery, and surgical oncology services so that the patient fully understands her options and their benefits and risks.

What should the initial definitive therapy for the cancer be?

Ductal carcinoma in situ

Surgical procedures

DCIS may be treated with breast conserving surgery or mastectomy with or without immediate reconstruction. The decision for type of surgery is based on the extent of the lesion relative to breast size, whether adequate clear margins can be achieved with partial mastectomy, ability to receive radiation therapy, and patient preference.

With mastectomy, skin sparing techniques can be utilized to facilitate reconstruction with excellent cosmesis. However, it is important that anterior margins of the mastectomy are clear of calcifications and tumor. for thin patients with large areas of microcalcifications, skin sparing may result in a positive margin on the mastectomy specimen if skin over these calcifications is not taken in the resection.

Radiation Therapy

For DCIS treated with breast conserving surgery, radiation is indicated in most circumstances to reduce risk of local recurrence. Clinical trials (including two large randomized studies, NSABP B17 and EORTC 10853) have demonstrated a 50-60% reduction in local recurrence. However, for certain tumors, that absolute benefit is low.

The following factors should be considered in determining the role of radiation therapy:

• size of the DCIS

• tumor grade and extent of comedonecrosis

• width of margins

• patient age

The Van Nuys index assigns points to each of these features to determine a total score ranging from 4-12 (see Table II). For patients with scores of 4-6, radiation does not appear to provide significant benefit, with local recurrence rates in this population of 1% at 5 years and 3% at 10 years. However, the index has not been validated in prospective randomized trials comparing radiation versus no radiation.

Table II.

The question of adding radiation to surgery has remained controversial. The NSABP B-17 trial randomized women with DCIS who had undergone partial mastectomy or lumpectomy with clear margins to no further treatment versus whole breast radiation.

When reported in 1993, the radiated group had a 7% local recurrence versus 16% for the non-radiated at a median follow-up of 43 months. Reduction in invasive recurrences was the most significant with 2.9% invasive recurrence versus 10.5% without radiation. In 1998, these results were updated at 8 years of follow-up.

At this point, the local recurrence rate was 27% for lumpectomy alone compared to 12% for lumpectomy with radiation. Fortunately, the highest persistent benefit of radiation was for reduction of risk for invasive cancer recurrences which were only 4% in the radiated group compared to 13% in the non-radiated group. However, there was no difference in overall survival between the two groups.

At 12 years of follow-up, local recurrence occurred in 32% of patient with lumpectomy alone versus 16% for lumpectomy with radiation. The investigators concluded that there was no subgroup that did not benefit from adjuvant radiation. Positive margins and the presence of comedonecrosis were associated with higher rates of local recurrence. In the EORTC 10853 trial (which included 1010 women with DCIS randomized to lumpectomy alone versus lumpectomy with radiation), the local recurrence rate at 10 years was 15% for lumpectomy with radiation compared to 26% for lumpectomy alone.

In the above described trials, radiation was administered to the whole breast over a period of 6 weeks. In the last decade, there has been interest in reducing the duration and extent of breast radiation with the use of partial breast radiation. In the pilot studies, DCI tumors were generally excluded, however, as techniques for accelerated partial breast irradiation (APBI) have evolved, there has been rapid adoption prior to the publication of randomized trials.

For this reason, the American Society for Radiation Oncology (ASTRO) issued recommended guidelines for use of accelerated partial breast irradiation. They defined 3 categories of patients: suitable, cautionary and unsuitable outside of a clinical trial. DCIS patients were not included in the “suitable” category. Patients with small (=3cm) DCIS lesions were included in the “cautionary” group, and patients with larger-size DCIS were categorized as “unsuitable”.

In response to these recommendations, investigators have reported their experience with APBI in the cautionary and unsuitable group. Jeruss et al reported the experience in the Mammosite Registry trial with DCIS lesions. These patients were treated with afterloaded radiation into a balloon catheter (Mammosite) over a period of 5 days. This delivers radiation to a limited field, approximately 1cm beyond the excision cavity. Among 194 patients with DCIS, with a median follow-up of 54 months, the actuarial 5-year incidence of local recurrence was 3.4%. Notably, nearly 50% of the DCIS lesions has comedonecrosis.

Surgical margins

The width of margin for breast conserving surgery is controversial. There is general agreement that less than 1mm is not ideal. However, the ideal margin width is less clear. Most surgeons prefer to achieve a 2mm radial margin. It is notable that margin width is considered more with DCIS than with invasive breast cancer. This is related to the growth pattern in a linear and branching pattern compared to an expanding invasive cancer mass.

In general, with resection margins greater than 10mm, there is a low recurrence rate, and absolute benefit of radiation is low. However, considerations of other factors such as young age, high tumor grade and large tumors give more imperative for radiation despite wide margins. Cases with margins under 1mm derive the greatest benefit from radiation.

A recent analysis suggests that patients with Van Nuys scores of 4-6 or 7 with a margin width above 3mm can be treated with excision alone and achieve local recurrence rates of below 6%. The addition of radiation achieves acceptable rates of local control (<20% local recurrence at 12 years) for those with score 7 and have margins under 3mm, patients who score 8 and have margins equal to 3mm, and for patients who score 9 and have margins equal to 5mm.

A meta-analysis of 4,660 patients treated with lumpectomy and radiation showed that a negative margin was associated with a substantially lower risk of local recurrence compared to positive margins. A margin of 2mm was superior for local control compared to under 2mm, with a relative reduction of local recurrence of 47%. There was no additional advantage to margins greater than 5mm.

Radiographic evaluation

When breast conserving surgery is undertaken for DCIS, the specimen should be examined with a specimen radiograph to confirm retrieval of the lesion of concern and all suspicious calcifications. The radiograph is obtained in two views to assess proximity of the mammographic lesion to the margins of resection.

If the lesion is near a margin, then re-excision of that margin can be performed at the first procedure. If there is a mass lesion or hematoma from the needle biopsy, a specimen ultrasound can be performed to assess margin width to the sonographic lesion. It is ideal to achieve clear margins at the first resection. This can be achieved by directed margin re-excision as described above or with routine shave cavity margins from all six margins.

Lymph node evaluation

While a non-invasive breast cancer should not have metastases to lymph nodes, there has been interest in the use of sentinel node biopsy to assess nodal status. In fact, studies have shown that sentinel node evaluation identifies nodal metastases in 5-12% of previous cases. In these cases, metastases were often identified by immunohistochemical staining and serial sectioning.

When metastases are identified by immunohistochemistry alone, axillary dissection may be overtreatment. Thus, it is not generally recommended for patients with DCIS found on needle biopsy to undergo sentinel node biopsy.

However, because invasive foci may be identified at the time of definitive resection, it may be necessary at that point to determine the nodal status. While sentinel node biopsy is accurate after a partial mastectomy, it is not possible to perform sentinel node biopsy after a mastectomy. Thus, performance of sentinel node biopsy is recommended for patients undergoing a mastectomy for DCIS.

Circumstances in which sentinel node biopsy should be considered even with breast conserving surgery include:

• suspicion of invasion on core needle biopsy

• large area of DCIS

• presentation with a mass lesion

• imaging findings suspicious for invasion

Endocrine therapy

The benefit of systemic therapy for DCIS has also been examined. The NSABP B-24 randomized 1804 women undergoing breast conserving surgery and radiation between either tamoxifen 20mg daily or placebo. This data was reported in 1999 with fewer breast cancer events in patients receiving tamoxifen compared to those who did not, 8.2% compared to 13.4%.

The risk of ipsilateral invasive recurrence was decreased from 4.2% to 2.1% at 5 years with tamoxifen. It should be noted that 16% of patients had positive or unknown margin status. These patients may derive the greatest benefit from tamoxifen. Unfortunately, in this trial patients were not selected by Estrogen receptor (ER) status, so benefits may have been diluted by the ER negative population.

Current guidelines recommend tamoxifen for ER with tumors treated with breast conserving surgery. The benefit of tamoxifen for ER negative tumors is uncertain. The side effect profile of tamoxifen and the low absolute differences in outcomes may impact whether a patient or her physician will include tamoxifen therapy after breast conserving surgery. Adverse side effects of tamoxifen include increased risk of thromboembolic phenomenon and uterine cancer.

The aromatase inhibitor anastrozole has been evaluated against tamoxifen in postmenopausal women in two trials (NSABP-B35 and IBIS-II DCIS), which demonstrated non-inferiority and an improved breast-cancer free interval, respectively. The benefit of an aromatase inhibitor is particularly apparent for postmenopausal women under the age of 60. Aromatase inhibitors offer a different side-effect profile, with an increase in fractures, musculoskeletal pain, and cerebrovascular events, but reduced risk of thromboembolic events, vasomotor symptoms or gynaecologic malignancies compared with tamoxifen. Trials of another aromatase inhibitor letrozole are currently ongoing.

Consideration can be given to risk reduction strategies for the contralateral breast in patients treated with mastectomy, although clinical trials have not specifically addressed this group of high-risk patients (see chemoprevention strategies below). The NCIC trial investigation of exemestane for breast cancer prevention included a small number of patients (2.5% of total) who had undergone mastectomy for DCIS.

The recommended doses of hormonal therapy for DCIS after breast conserving surgery:

• Tamoxifen 20 mg PO daily for 5 years in premenopausal or postmenopausal women

• Anastrozole 1 mg PO daily for 5 years in postmenopausal women

Lobular carcinoma in situ

Any “treatment” of LCIS must take into account that the LCIS confers risk for breast cancer in both breasts. Local resection after a core needle biopsy showing LCIS is warranted to confirm the diagnosis of LCIS only and to rule out DCIS or invasive cancer.

In a review of multiple published studies of surgical excision after a needle biopsy revealing LCIS, DCIS or invasive cancer was identified in 27% of cases. However, radiation post excision is not indicated. If surgical therapy is undertaken, the procedure should be bilateral prophylactic mastectomy with or without immediate reconstruction. A decision for bilateral mastectomy is most likely to be made when there are other risk factors such as a significant family history of breast cancer or known BRCA mutation.

While classical LCIS is not managed with directed local therapy and radiation, the pleomorphic LCIS variant may require more directed therapy. Because this variant seems to behave more like DCIS, directed local therapy should be considered. Margin clearance with partial mastectomy should be achieved with consideration for adjuvant radiation. There is need for more data on the natural history of this variant.

The medical approach to LCIS is that of chemoprevention. Randomized clinical trials of chemoprevention in women at high risk for breast cancer have shown reduction in risk of invasive breast cancer for women with LCIS. In the NSABP Breast Cancer Prevention Trial-BCPT, tamoxifen reduced the risk of invasive cancer by 50% compared to placebo in those whose increased risk was due to LCIS.

In the subsequent STAR trial, comparing tamoxifen to raloxifene in high-risk women, raloxifene was also found to be efficacious. However, in longer term follow-up, raloxifene was only 75% as effective as tamoxifen in reducing risk of breast cancer.

In 2011, a study of chemoprevention with an aromatase inhibitor, exemestane was reported. In a median follow-up of 3 years, breast cancer events were reduced by over 50% with exemestane compared to placebo. The risk of invasive breast cancer was reduced by 65%. The overall side effects were similar between placebo and exemestane groups. Statistically there were significantly more complaints of hot flashes and arthritis with exemestane. This trial included patients with atypical hyperplasia and LCIS, accounting for 8% of participants.

Tamoxifen is the only approved agent for breast cancer prevention in pre-menopausal women. Raloxifene and exemestane were only evaluated in post-menopausal women.

The recommended dosages for these agents are:

• Tamoxifen 20 mg PO daily

• Raloxifene 60 mg PO daily

• Exemestane 25 mg PO daily

In clinical trials these prevention agents were administered for 5 years.

What should you tell the patient and the family about prognosis?

Ductal carcinoma in situ

DCIS is curable and the prognosis is excellent. Because it is non-invasive, the tumor does not have the capacity to spread distantly. In rare cases, particularly with large DCIS lesions greater than 4cm, metastatic disease to lymph nodes or distant sites may occur. This is likely due to unidentified areas of micro-invasion that are not seen during pathologic sampling.

The risk of recurrence of in situ or invasive cancer in the ipsilateral breast is related to the extent of surgical treatment:

• less than 1% with total mastectomy

• 7-11% at 10 years with breast conserving surgery and radiation.

For patients with Van Nuys scores of 4-6, recurrence risk is 6% or less. The most serious adverse consequence of recurrence is that the lesion may be invasive 50% of the time when it recurs, which then exposes the patient to the risk of distant disease.

Lobular carcinoma in situ

LCIS is a risk factor for the development of subsequent breast cancer in either breast. The risk of subsequent invasive breast cancer ranges from 7% to 17% at 10-15 years after LCIS diagnosis. Slightly more than half of these cases occur in the contralateral breast. This risk can be reduced by endocrine systemic therapy, with tamoxifen for pre-menopausal women, and tamoxifen, raloxifene or exemestane for post-menopausal women.

What if scenarios.

Error in pathologic diagnosis

In some cases, it can be difficult to differentiate LCIS and DCIS. E-cadherin staining by the pathologist can help to make this discrimination. Second opinion pathology review is sometimes needed to clarify the diagnosis.

Failure to retrieve all microcalcifications with a partial mastectomy (Lumpectomy)

The risk of this can be reduced by the performance of an intraoperative radiograph of the specimen and view with magnification. If calcifications are near the edge of the specimen, then it is prudent to re-excise that margin at the first procedure.

It is also a good idea to get a post-operative mammogram to confirm there are no remaining lesional microcalcifications at the operative site. If residual calcifications are seen, these can be wire localized and resected.

Positive margin on total mastectomy specimen

This usually occurs with extensive DCIS that is close to the skin, particularly in thin patients. It is more likely to occur with skin sparing mastectomy. To avoid this, it is possible to x-ray slices of the mastectomy specimen to confirm that calcifications are not seen near the surface of the specimen.

This approach can be tedious, so careful surgical planning will be more likely to be successful. This may involve sacrificing skin overlying the lesion. Another strategy is sending pathology samples from the elevated skin flap overlying the lesion to confirm no breast tissue or tumor is remaining on the skin flap.

Follow-up surveillance and therapy/management of recurrences.

Ductal carcinoma in situ

For patients with DCIS treated with breast conserving surgery with or without radiation, follow-up consists of:

• Breast exam every 6 months.

• Mammography of the ipsilateral breast every 6 months for 3 years to assess changes from surgical scarring and radiation. When the changes have stabilized, mammography intervals can be yearly.

For patients with DCIS treated with breast conserving surgery, adjuvant tamoxifen should be considered to reduce risk of local recurrence in the conserved breast. If recurrence develops in the breast after breast conserving surgery and radiation, then the patient must undergo mastectomy with or without immediate reconstruction.

If the recurrence is non-invasive, then adjuvant systemic therapy is not warranted. However, if the tumor is invasive, then it would be treated as appropriate for stage of disease and tumor profile.

Lobular carcinoma in situ

For patients who elect observation with or without chemoprevention, enhanced surveillance is planned. This includes physical exam and breast imaging at 6 month intervals. Bilateral mammograms should be performed annually.

MRI of the breasts may be utilized alternating with mammogram at 6 month intervals, particularly, for women with dense breast tissue. American Cancer Society guidelines for breast screening in women at high risk for breast cancer recommend the addition of MRI exams to mammographic screening for women with at least a 20-25% lifetime risk of breast cancer.

NCCN guidelines also recommend considering breast MRI screening in this circumstance. A study of enhanced screening with MRI in patients with LCIS showed that 4% of patients had cancer on MRI with a negative mammogram.

If a patient develops an invasive cancer, lobular or ductal after a diagnosis of LCIS, she is a candidate for either breast conserving surgery or mastectomy depending on the nature and size of the breast cancer.

Pathophysiology

Ductal carcinoma in situ

DCIS refers to malignant ductal cells that are confined in the ducts of the breast and have broken through the basement membrane to become invasive. Thus, these lesions do not have metastatic potential, but they do have the potential to become invasive.

The histology may vary including cribriform (Figure 5), papillary (Figure 6), solid (Figure 7) and comedo (Figure 8) types. The cribriform, papillary and micropapillary are more likely to be low grade, while the solid and comedo type are more likely to be high or intermediate grade. Histologic grade is assigned on the pathology exam as high, intermediate or low. Grade impacts risk for local recurrence and increased probability of invasion being found in the definitive surgical resection. Alternatively, lesions may be graded as high or non-high grade with or without necrosis.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

DCIS lesions have been evaluated to assess common tumor markers associated with invasive breast cancers. Estrogen receptor (ER) and progesterone receptor (PR) positivity has been observed in 49–97% and 40–83%, of tumors, respectively. Her2neu positivity has been observed in approximately 40% of tumors. Factors associated with HER-2 positive DCIS include extensive DCIS, Grade III, estrogen and progesterone receptor negative, central necrosis, comedo subtype, and post-menopausal status.

The natural history of in situ lesions is important to understand in order to design intelligent approaches to therapy. If DCIS is not surgically treated, there is significant risk of developing an invasive cancer in the area of the DCIS. This is in contrast to LCIS which has risk for the development of invasive breast cancer in either breast over time. In summary, LCIS is considered a risk factor for invasive cancer while DCIS is considered a precursor to invasive cancer.

The natural history of DCIS treated by excision alone without radiation has been reported in several studies with long follow-up. Lagios reported on 79 women with DCIS smaller than 25mm treated with wide excision only. At 124 month of follow-up, there was a 16% rate of local recurrence and 50% of recurrence was invasive. Local recurrence varied significantly with a 33% rate of local recurrence when high grade with comedonecrosis versus 2% in intermediate or low grade DCIS.

Hetelekidis reported a series of 59 patients with a follow-up of 95 months with a 10% local recurrence at 5 years, 40% of which were invasive. Lesion size was the most important factor in local recurrence.

Lobular carcinoma in situ

LCIS arises in the terminal duct lobular unit and may have pagetoid spread into the ducts (Pagetoid spread refers to lobular cells trailing into the duct from the lobule), the basement membrane is intact and the cells fill more than half of the lumen.

Pleomorphic LCIS (PLCIS) is a variant of LCIS which is characterized by large nuclei and prominent nucleoli and more mitotic figures than are commonly seen in classic LCIS. Central necrosis may be seen similar to comedo necrosis in ductal carcinoma in situ. Its histologic similarity to ductal carcinoma in situ can create confusion in diagnosis.

It can be difficult to differentiate PLCIS from DCIS on pathology exam. E-cadherin immunohistochemical stains will be negative with PLCIS and positive with DCIS. It is more likely that the mammogram will show microcalcifications associated with PLCIS than with classic LCIS (Figure 9).

Figure 9.

What other additional laboratory studies may be ordered?

Extensive laboratory work-up is not warranted for DCIS and LCIS since they do not have capacity for systemic spread.

Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.

No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.