Radiation Oncology: A Question-Based Review

Ductal and Lobular Carcinoma In Situ

John P. Christodouleas and Atif Khan

image Background

Ductal carcinoma in situ (DCIS) represents what % of all breast malignancies?

DCIS represents ~25% of all breast malignancies.

Which is more common: DCIS or lobular carcinoma in situ (LCIS)?

DCIS is 5 times more common than LCIS.

Name the 5 most common histologic subtypes of DCIS.

Most common subtypes of DCIS:

1.     Cribriform

2.     Comedo

3.     Papillary

4.     Medullary

5.     Solid

(Mnemonic: C2PMS)

Which histologic subtypes of DCIS have the worst and 2nd worst prognosis?

The DCIS subtype that has the worst prognosis is comedo, and the 2nd worst is solid. DCIS is often grouped into comedo and noncomedo subgroups.

What is the most common clinical presentation of DCIS?

DCIS most commonly presents with microcalcifications on a mammogram.

What is the most common clinical presentation of LCIS?

LCIS most commonly presents as an incidental finding. LCIS typically does not result in mammographic or clinical abnormalities.

What is the rate of progression at 10 yrs of DCIS to invasive Dz if left untreated?

~30% of DCIS progress to invasive Dz at 10 yrs if left untreated. (Page DL et al., Cancer 1995)

For a pt with LCIS, what is the risk of the pt to be diagnosed with invasive Dz by 10 yrs?

A pt with LCIS has an ~7% risk of developing invasive cancer at 10 yrs (~1%/yr), but approximately half of invasive Dz occurs at the contralat breast, suggesting that LCIS is probably just a marker for propensity to form invasive Dz. (Chuba PJ et al., J Clin Oncol 2005)

What % of pts with LCIS who subsequently develop invasive Dz develop invasive lobular cancers?

Only 25%–50% of subsequent cancers are invasive lobular cancers (i.e., though LCIS is a proliferative lesion of the lobules, it is mostly a marker for subsequent ductal proliferative lesions).

For a pt with LCIS, what is the risk of invasive Dz in the ipsi breast vs. the contralat breast?

For a pt with LCIS, the risk of subsequent invasive Dz is equal in both breasts.

How many pathologic grades are there for DCIS?

There are 3 pathologic grades for DCIS: low, intermediate, and high.

What % of DCIS are estrogen receptor (ER)+?

75%–85% of DCIS cases are ER+.

Which subtype of LCIS has the worst prognosis?

Of LCIS subtypes, pleomorphic LCIS has the worst prognosis.

image Workup/Staging

What is the initial workup after a DCIS Dx?

DCIS workup: H&P (with emphasis on risk of hereditary breast cancer), diagnostic bilat mammogram, assessment of ER status, +/− genetic counseling

Is an axillary dissection needed for DCIS?

No. An axillary dissection is not needed for DCIS. However, per NCCN 2010, consider if (a) the pt is undergoing mastectomy for Tx or (b) if the location of lumpectomy will compromise future sentinel Bx should it be necessary.

What is the T stage for DCIS?

DCIS has its own designation: Tis.

What is the definition of DCIS with microinvasion, and what is the significance for workup?

DCIS with microinvasion refers to invasion >1 mm in size. If microinvasion is present, then a sentinel LN Bx is indicated, as the LNrate is ~4%–8%.

For a pt with DCIS, if there is <1-mm margin at excision, what is the rate of residual Dz at the time of re-excision?

For a pt with DCIS and a <1-mm margin at excision, ~30% will have residual Dz at re-excision. Notably, low- and intermediate-grade DCIS is more likely to grow in a discontinuous pattern (Faverly DR et al., Semin Diagn Pathol 1994). B/c of this, margin status may be, paradoxically, more important in these lesions. In these discontinuous type lesions, gaps of uninvolved tissue between DCIS are typically small (<5 mm in 80% of cases).

For a pt with DCIS, in which situation would re-excision not be indicated with a margin <1 mm?

If a pt with DCIS has an excisional margin <1 mm at the fibroglandular border of the breast (skin or chest wall), then re-excision is not indicated.

image Treatment/Prognosis

What is the Tx paradigm for unifocal DCIS?

There are 2 Tx paradigms for unifocal DCIS:

1.     Lumpectomy + PORT +/− tamoxifen (if ER+)

2.     Mastectomy

What must be done after lumpectomy for DCIS to ensure that all the Dz has been removed?

Post-excision mammography to ensure that all the microcalcifications are removed. Specimen radiograph is done after lumpectomy to ensure that the calcifications are removed.

For a pt with DCIS, what is rate of LR after mastectomy alone?

For a pt with DCIS, the rate of LR after mastectomy is ~2% at 10 yrs.

What are considered adequate surgical margins in pts receiving breast conservation surgery for DCIS?

A systematic review of published trials in DCIS with breast conservation therapy (BCT) involving 4,660 pts found that a 2-mm margin was superior to a margin <2 mm (OR 0.53), without any LC benefit in margins >2 mm. (Dunne C et al., J Clin Oncol 2009)

What are the contraindications for BCT for DCIS?

Contraindications for BCT for DCIS: multifocal Dz, persistently +margins, cosmetic limitations, inability to get postop RT (pregnancy or prior RT).

For a pt with DCIS treated with lumpectomy, what is the impact of PORT on ipsi breast recurrence (invasive and noninvasive) and OS?

For DCIS treated with lumpectomy, PORT reduces LR by 50%, but there is no evidence for OS benefit.

Is there a benefit of mastectomy over BCT for DCIS?

This is undetermined. No prospective study has directly compared mastectomy vs. BCT for DCIS. However, in the NSABP B06 study, 78 pts were found to have DCIS after final pathology review. The LF rate was 2% for mastectomy, 9% for lumpectomy + RT, and 43% with lumpectomy alone (Fisher B et al., JSO 1991).

Name 4 prospective studies that support the addition of RT after lumpectomy in pts with DCIS.

Prospective studies that show LC benefit after lumpectomy in DCIS:

1.     NSABP B17 (Fisher B et al., Sem Oncol 2001)

2.     EORTC 10853 (Bijker N et al., J Clin Oncol 2006)

3.     UKCCCR (Houghton J et al., Lancet 2003)

4.     SweDCIS (Holmber L et al., J Clin Oncol 2008)

Describe the Tx arms and the invasive and noninvasive LR outcomes in NSABP B17 and EORTC 10853.

1.     In NSABP B17, 818 DCIS pts with –margins (no cells at inked margin) treated with lumpectomy were randomized to 50 Gy ipsi RT or no RT. At 10 yrs, the overall ipsilateral breast tumor recurrence (IBTR) rate was 30.8% vs. 14.9% in favor of RT. Both the invasive and noninvasive recurrence rate was approximately halved by RT.

2.     In EORTC 10853, 1,010 DCIS pts with –margins (no cells at ink) were randomized to lumpectomy vs. lumpectomy + whole breast RT at 50 Gy. RT reduced LF from 16% to 9%. Half of all recurrences were invasive.

For a pt with ER+ DCIS, is there a benefit to tamoxifen in addition to lumpectomy + RT? What studies support this?

1.     The evidence for the benefit of adj tamoxifen in addition to lumpectomy + RT is mixed.

2.     NSABP B24 compared lumpectomy + RT +/− tamoxifen and found that at 7 yrs, the addition of tamoxifen significantly decreased IBTR (8% vs. 11%). The benefit was seen with respect to invasive recurrences but not noninvasive recurrences. CBTR was also reduced with tamoxifen (2.3% vs 4.9%) (Fisher B et al., Semin Oncol 2001). 2 caveats about this trial are that –margins were not required (the inclusion of margin+ or margin unknown pts may have exaggerated the benefit of tamoxifen) and that index lesions were not tested for ER status at the time, though subsequent analyses have attempted to address this.

3.     Another randomized trial by the UKCCCR (Houghton J et al., Lancet 2003) did not show a benefit with the addition of tamoxifen with respect to either invasive or noninvasive recurrences. Differences between these studies may be due to the age of cohorts, as NSABP B24 enrolled generally younger women, and younger age at Dx was shown to be associated with higher IBTR rates in the B24 report. In B24, 30% of pts were <50 yo. In UKCCCR, 10% of pts were <50 yo. ER status has not been compared between the 2 studies. In addition, the design of the UKCCCR trial (2 × 2 factorial design) and institutional choice in participating in 2 × 2 design or only 1 of the randomizations may have created imbalances in the Tx arms. In summary, it is controversial which subgroups of DCIS pts treated with lumpectomy truly benefit from adj tamoxifen, but in current practiceERDCIS pts are offered adj tamoxifen.

For a pt with DCIS, what is the effect of adj tamoxifen on contralateral breast tumor recurrence (CBTR)?

NSABP B24 showed that the addition of tamoxifen to BCT in DCIS pts significantly reduced CBTR as the 1st site of recurrence from 4.9% to 2.3% at 7 yrs.

For a pt with ER– DCIS, is there benefit to adj tamoxifen after lumpectomy + RT?

No. Retrospective analysis of NSABP B24 showed that the benefit of adj tamoxifen was limited to ER+ pts. (Allred DC et al., Br Cancer Res Treat 2002)

For a pt with ER+ DCIS, does adj tamoxifen obviate the benefit of RT after lumpectomy? What study evaluated this?

RT is still beneficial for pts with DCIS treated with lumpectomy even with adj tamoxifen. UKCCCR was a 2 × 2 factorial study looking at the benefit of RT and tamoxifen in DCIS pts after lumpectomy. After a median follow-up >4 yrs, RT reduced IBTR in women given adj tamoxifen (6% vs. 18%). (Houghton J et al., Lancet 2003)

For a pt with DCIS, name some risk factors associated with LR and which is most important.

Risk factors for LR in a pt with DCIS:

1.     Decreased margin width (most important)

2.     Increased size of tumor

3.     High grade

4.     Young age (≤40 yrs)

5.     Comedo necrosis

6.     Multifocality

For a pt with DCIS treated with lumpectomy + PORT (no tamoxifen), estimate the risk of LR if the pt had a negative surgical margin vs. a positive surgical margin.

For a pt with DCIS treated with lumpectomy + PORT (no tamoxifen), LR at 14 yrs is ~13% if a –margin vs. 26% if a + margin.

What is the purpose of the Van Nuys Prognostic Classification, and what are its limitations?

The purpose of the Van Nuys Prognostic Classification system is meant to identify DCIS pts who are at low risk for recurrence after RT alone using width of margins, size, grade, and age. The system was developed retrospectively and has not been validated in prospective studies or in different retrospective data sets. (Silverstein MJ et al., Am J Surg 2003)

Do all DCIS pts require PORT?

1.     This is controversial. Some data suggest that all women with DCIS need PORT. To date, the prospective randomized trials have failed to identify a subset of DCIS pts who can safely omit adj RT. The Van Nuys investigators have published retrospective data showing low IBTR rates in a subset of selected DCIS pts using special surgical and pathologic techniques. Other groups have not been able to reproduce these results. However, recent unpublished data from the ECOG suggest that recurrence is low without PORT for low- to intermediate-grade DCIS excised to widely –margin (5–10 mm). For high-grade Dz +/− comedo necrosis, PORT is necessary.

2.     The DFCI trial was a prospective study that enrolled women with DCIS after surgery to >1-cm margin, grade 1–2 Dz, to undergo observation. Accrual was halted after 5 yr IBTR reached 12%, meeting early stopping rules of excessive failures. The trial, however, assessed mammographic size (rather than pathologic size) as an entry criteria, did not allow tamoxifen, and included women with “predominant grade 1 or 2 Dz,” implying that higher-grade components may have been present in some cases. Nonetheless, the results of the trial are in keeping with the body of literature in general. (Wong J et al., J Clin Oncol 2006)

3.     ECOG E5194 (Hughes LL et al., J Clin Oncol 2009): prospective study, 711 pts with either low- to intermediate-risk (low/intermediate-grade Dz, ≤2.5 cm) or high-risk (high grade, ≤1 cm) DCIS were excised to >3-mm margins. All nonpalpable Dz was ≥3 mm. No PORT. Tamoxifen was taken by 31%. Accrual to the low/intermediate-grade group was separate from the high-grade cohort. The majority had 5–10-mm margins. 606 pts were in the low/intermediate group, and 105 pts (stopped due to poor accrual) were in the high-grade group. Median follow-up was 6.2 yrs, median age was 60 yrs. Results: Low/intermediate-grade cohort: 5-yr and 7-yr IBTR was 6.1% and 10.5%; CBTR was 3.7% and 4.8%. For the high-grade cohort: 5-yr and 7-yr IBTR was 15.3% and 18%; CBTR was 3.9% and 7.4%. There was 50% recurrence with invasive Dz. These results clearly support PORT in even the smallest high-grade DCIS pts, but longer follow-up is warranted to determine whether omission of RT is appropriate for the low-risk groups. Retrospective data sets have demonstrated that with longer follow-up, lower-grade DCIS pts have similar IBTR rates as higher-grade pts.

4.     RTOG 98-04: RCT with the same selection criteria as the ECOG study randomized 636 pts to RT vs. no RT. Results are pending.

What whole breast dose and boost dose is used for a pt with DCIS after lumpectomy?

For DCIS, the whole breast dose is 46–50 Gy with a lumpectomy bed boost to a total of 60–66 Gy. The role of boost in DCIS is controversial. The practice is extrapolated from results of the EORTC and Lyon boost trials for invasive cancers, but there have been no prospective trials of the role of boost in DCIS pts. ~44% of pts from B24 were given a boost, mainly for +margins.

What is the Tx paradigm for LCIS?

LCIS Tx paradigm: For pure LCIS after lumpectomy, pts can be observed +/− risk reduction procedures. Occurrence of invasive Dz after LCIS is low and is often in the contralat breast. In addition, invasive Dz after LCIS is generally relatively favorable, and deaths subsequently are rare. The exception may be pleomorphic LCIS, though data are limited.

For a pt with LCIS, what are 2 options to reduce the risk of development of an invasive cancer?

Options to reduce the risk of development of an invasive cancer:

1.     Antiestrogen therapy with tamoxifen or raloxifene (raloxifene only if postmenopausal) (NSABP P1 trial)

2.     Bilat mastectomy +/− reconstruction if there is a risk of hereditary breast cancer

What is the management for a woman with LCIS detected on percutaneous core needle Bx?

LCIS does not result in mammographic or physical exam findings; as such, pure LCIS on core needle Bx should typically prompt an open Bx to identify the reason for the abnl mammographic findings that prompted the initial Bx.

Is LCIS associated with invasive cancer a contraindication for BCT?

No. Current literature supports the safety of BCT in the presence of coexisting LCIS in the specimen, and no special effort needs to be made to obtain –margins on LCIS.

For a pt with LCIS, what is the benefit of primary tamoxifen?

For a pt with LCIS, tamoxifen halves the risk of invasive recurrence in either breast. (Fisher B et al., Lancet 1999)

In a pt with DCIS or invasive Dz, what is the most common contraindication for adj tamoxifen therapy?

In a pt with DCIS or invasive Dz, the most common contraindication for adj tamoxifen therapy is Hx of stroke or other coagulopathy.

image Toxicity

What is the recommended follow-up schedule after Tx for DCIS?

Recommended follow-up schedule after Tx for DCIS: interval H&P exam every 6–12 mos for 5 yrs, then annually. Mammogram for the treated breast every 6–12 mos and annually for the contralat breast.

For a pt with LCIS, what is the recommended observation strategy?

Recommended observation strategy for a pt with LCIS: H&P every 6–12 mos and bilat mammogram annually (NCCN 2010)