Clinical Challenges: The Emerging Identity of Lobular Breast Cancer

During a year-in-review session at the 2022 San Antonio Breast Cancer Symposium (SABCS), a summary of presentations on lobular breast cancer described the subtype as “the Unseen.” Fast forward 1 year, and SABCS devoted an entire session to lobular disease.

“That [2022 presentation] really spoke to me,” said Jason Mouabbi, MD, of the University of Texas MD Anderson Cancer Center in Houston, during this year’s SABCS education session on lobular breast cancer. “During this conference, we’re seeing a lot more ILC [infiltrating lobular carcinoma] abstracts, between 25 and 30 in this conference, and the number keeps growing with every conference. That’s great to see.”

Chair of the scientific advisory board of the Lobular Breast Cancer Alliance, Mouabbi has seen signs of growing recognition of the similarities and differences between lobular and ductal breast cancer. His hope is that with increasing awareness, the nuances of diagnosis and treatment of ILC will figure more prominently into clinical management.

ILC accounts for 10-15% of newly diagnosed breast cancers. The hallmark histopathologic difference between ILC and infiltrating ductal carcinoma (IDC) is loss of the anchoring cell surface protein E-cadherin, most often due to a mutation in CDH1. A similar proportion of ILC and IDC is stage I-II at diagnosis, but about 90% of ILC is grade 1-2 at diagnosis as compared with 60% of IDC. Almost twice as many ILCs have low proliferation activity as assessed by Ki-67 (60% vs 35%). Consistent with breast cancer in general, most ILCs are estrogen receptor (ER) positive.

ILC is associated with worse long-term outcomes as compared with IDC, even though ER-positive IDC is six times more likely to be high risk as determined by a 21-gene risk-recurrence assay. A possible contributing factor is that ILC is more likely to develop visceral metastases.

Another factor in the worse long-term outcome with ILC relates to the impact of ER status. Late recurrences are more common in ER-positive tumors, regardless of histology, noted Tari King, MD, of Dana-Farber/Brigham and Women’s Cancer Center in Boston. However, an analysis of disease-free survival (DFS) and overall survival (OS) showed inferior outcomes for ILC, regardless of ER status.

“The same relationship holds true and the shape of the [survival] curves are similar for the ER-positive and ER-negative patients, suggesting that this time-dependent effect on prognosis is actually independent of ER status,” said King.

Rates of local recurrence after breast-conserving therapy or mastectomy are similar between ILC and IDC. Regional recurrence is more common in IDC, whereas recurrence in bone and distant recurrence are more frequent in ILC, King noted.

Considerations for Imaging

Conventional breast imaging modalities — mammography and ultrasound — have poor sensitivity (60-70%), high false-negative rates (30-40%), and poor size concordance with pathology (32-45%) for ILC, said Mouabbi. In dense breast tissue, mammography sensitivity for detecting ILC falls as low as 11%.

Multiple studies have evaluated the correlation between ILC tumor size on imaging and pathologic tumor size. The studies have consistently shown that MRI has the best correlation, said King. However, MRI, like other types of imaging, has its shortfalls. For example, one small single-center series showed only 56% correlation between estimated tumor size on MRI and tumor size on pathology. In particular, MRI overestimated the size of ILC in almost a third of cases.

“Keep in mind that most MRIs are diagnostic MRIs [as opposed to screening], so we already know that there is a cancer,” said Mouabbi. “The sensitivity is about 93% and the false-negative rate is low, about 7%. What pops out is the false-positive rate, and we see that a lot in clinical practice. We see false positives in about a third of cases.”

The clinical implications of breast MRI accuracy for ILC were evaluated in a meta-analysis of nine studies. Among other findings, the analysis showed that preoperative MRI was associated with higher rates of mastectomy, overall and in the ILC subgroup. The investigators also found that MRI was associated with lower rates of re-excision in ILC.

The clinical picture had changed by the time of an update to the meta-analysis. The association between MRI and upfront mastectomy had disappeared, and the association with lower re-excision rates in ILC was no longer statistically significant.

“Despite the tendency toward increased use of MRI in these patients, we have yet to be able to measure improvement in surgical outcome,” said King.

Contrast-enhanced mammography has shown promise for achieving diagnostic accuracy that is at least comparable to breast MRI, said Mouabbi. He described a typical, real case: a middle-age woman had a palpable mass in her right breast for 3 years. Serial mammography and ultrasound were either negative or unchanged from the prior year. A biopsy during one year also was negative.

“Then we decided to do contrast-enhanced mammography and we see all this enhancement, a multicentric ILC,” he said. “So contrast-enhanced mammography has great potential for ILC. The sensitivity is about the same as MRI. It still has some false-negative results but much lower than with [standard] mammography, about 22%. It still has some false positivity but way less than MRI.”

Treatment Considerations

The approach to treating ILC has changed dramatically since the subtype was characterized in the 1940s, said King. For decades, mastectomy remained standard of care until new information emerged about the disease’s natural history. Beginning in the 1990s and continuing into the next decade, “we became more comfortable counseling women about their risk, as opposed to performing a mastectomy.”

The decline in use of mastectomy, even for lobular carcinoma in situ (LCIS), gave way to excision after core biopsy. The most recent data no longer support routine excision of LCIS following core biopsy.

“Studies demonstrated that, with careful radiographic pathological correlation, excluding cases with additional lesions that would otherwise independently require excision, the rates of upgrade to invasive cancer are in fact quite low,” said King. “Many of us have moved away from routine excision for these lesions. We now have data on the safety of omitting routine excision.”

Data from King’s center documented a low risk of recurrence with observation following detection of LCIS or atypical lobular hyperplasia (ALH). Of 127 patients with LCIS or ALH, 25 underwent excision and the remaining 102 opted for observation. After a median follow-up of 3 years, the 3-year actuarial rate of invasive cancer was 3.9% in the patients who were observed.

“Importantly, there were no same-site cancers,” she said. “Of the five cancers that developed, two were ipsilateral but in a different quadrant and three were in the contralateral breast.”

LCIS is a risk factor for subsequent development of ILC, King continued. The 10-year cumulative risk for ILC after LCIS is 11%. Most women with LCIS develop early-stage breast cancer, and invasive cancers are mostly grade I or grade II. Chemoprevention with a selective ER modulator or aromatase inhibitor can reduce the risk of breast cancer after LCIS by 50%.

Multiple studies have evaluated strategies to de-escalate treatment of breast cancer, particularly studies of radiation therapy protocols. Patients with ILC have not been uniformly excluded from radiation de-escalation trials but they have been underrepresented, said King. Trials of partial breast irradiation (PBI) did exclude ILC, so PBI is not recommended for patients with this breast cancer subtype.

Neoadjuvant chemotherapy achieves lower rates of pathologic complete response (pCR) in ILC as compared with IDC, said King. An analysis of data for 9,020 patients (1,051 with ILC) from nine trials of neoadjuvant chemotherapy showed a pCR rate of 6.2% for ILC and 17.4% for IDC. Moreover, pCR in ILC was not associated with survival.

Several studies have failed to show a benefit for chemotherapy in patients with ILC. Most recently, a Dutch study of 716 patients with ER-positive ILC showed that three-fourths of the patients had an indication for chemotherapy, in addition to endocrine therapy, and three-fourths of those patients (n=379) received chemotherapy (neoadjuvant or adjuvant). After a median follow-up of at least 5 years, investigators found no difference in relapse-free survival, breast cancer-specific survival, or OS between patients who received adjuvant chemotherapy and those who did not.

Some newer approaches to treating ILC have yielded promising results, said Mouabbi. One recent study showed similar survival for patients with early ILC and IDC treated with endocrine therapy plus a CDK4/6 inhibitor.

Several novel approaches to targeting HER2-mutated breast cancer also have shown promise. An evaluation of neratinib (Nerlynx) plus fulvestrant with or without trastuzumab (Herceptin) showed a response rate of about 40% in patients with metastatic ILC or IDC. Duration of response, clinical benefit rate, and median progression-free survival were also similar.

Trials of therapy targeting the PTEN-PI3K-AKT pathway and ROS1 in advanced breast cancer have already begun. A good rationale exists for targeting DNA damage response in ILC with PARP inhibitors, as well as targeting the IL-6/STAT3 pathway, said Mouabbi.

A subgroup of ER-positive ILCs can be characterized as “immune related,” he continued. In the KEYNOTE-028 trial of pembrolizumab (Keytruda) in metastatic PD-L1-positive, ER-positive breast cancer, two-thirds of objective responses occurred in patients with ILC.

Genomic assays to inform treatment decisions have been shown to be prognostic for lobular cancers. However, the percentage of patients classified as high risk varies from one assay to another and among different breast cancer subtypes.

“Although we have a lot of experience with the genomic assays for lobular cancer, we’re probably going to need a genomic assay that’s specific to these cancers,” said King.

Toward that end, an ongoing study aiming to develop an ILC-specific biomarker panel has whittled an initial 194-gene signature down to expression of four key proteins. A report at SABCS showed that the assay is highly prognostic for lobular cancers as compared with non-lobular cancers.

Please enable JavaScript to view the

comments powered by Disqus.