Fellows Present Their Findings on the Breast Cancer Immune Microenvironment and Telehealth Delivery at National Fellows Forum

Tess O’Meara, MD, MHS

Beyond the exciting clinical trial updates presented at the 2024 San Antonio Breast Cancer Symposium (SABCS) in Texas, oncology fellows from across the US came together the day before the start of the meeting to share their research interests and network with other fellows and senior faculty members.

During the OncLive National Fellows Forum® at SABCS, which took place on December 9, 2024, Tess O’Meara, MD, MHS, and Kelsey H. Natsuhara, MD, presented findings from their studies on profiling tumor-associated immune cells in breast cancer and the impact of telehealth on patient access and treatment use. O’Meara is a fellow of breast medical oncology at Dana-Farber Cancer Institute/Massachusetts General Hospital in Boston, Massachusetts, and Natsuhara is a hematology/oncology fellow at the University of California, San Francisco (UCSF).

Describing the Immune Microenvironment in Breast Cancer

O’Meara’s presentation, titled “ImmunoPROFILE: Profiling Tumor- Associated Immune Cell Populations in Primary and Metastatic Breast Cancer Tissue,” looked to characterize the immune microenvironment of primary hormone receptor (HR)–positive/HER2-negative breast cancer relative to other breast cancer types via multiplexed immunofluorescence.1 The study included tissue samples from patients with primary and metastatic HRpositive/ HER2-negative breast cancer (n = 145), HER2-positive breast cancer (n = 20), and triple-negative breast cancer (TNBC; n = 36). Investigators evaluated immune cell infiltration in terms of CD8, PD-1, FOXP3, and CD8 and PD-1 cell positivity, as well as PD-L1 combined positive score, tumor proportion score, and inflammatory score.

“ImmunoPROFILE is a pan-cancer effort at Dana-Farber to use archival formalin-fixed paraffin-embedded tissue from primary and metastatic tumors to characterize the immune cell populations that are infiltrating those tumors,” O’Meara explained in an interview with Oncology Fellows. “My project focused on breast cancer, mostly HR-positive tumors with smaller cohorts of HER2-positive [breast cancer] and TNBC.”

Findings from the study demonstrated that PD-L1 levels per immunohistochemistry staining were correlated with PD-L1 by multiplexed immunofluorescence, supporting the use of the technique in the clinical assessment of immune cell populations. When comparing samples from patients with primary HR-positive/HER2-negative breast cancer (n = 116) with those with metastatic disease (n = 29), PD-1–positive T cells (P = .048) and PD-L1 combined positive score (P = .036) were significantly elevated in those with primary disease. There was also a trend toward higher immune cell staining reported in primary vs metastatic TNBC.

Additional multiplexed immunofluorescence data showed that there was decreased tumoral immunity in primary HR-positive/HER2-negative breast cancer compared with primary HER2-positive breast cancer or TNBC. There was also increased PD-1–positive immune cell infiltration in HR-positive/HER2-negative breast cancer in samples from patients older than 65 years compared with those who were younger than 45 years.

“Our most interesting and relevant findings came when we [examined] the primary HR-positive/HER2- negative cohort of cases: how immune cell infiltration differed with tumor [pathologic] grade,” O’Meara said. “We found that there was a stepwise increase of immune cell infiltration across all the populations we looked at with increasing [tumor] grade. With grade 1 cancers, there was the lowest degree of immune cell infiltration. There was somewhat higher infiltration in grade 2, and the highest immune cell infiltration [was observed] in grade 3. This is particularly relevant because the recently published [phase 3] KEYNOTE-756 study [NCT03725059] looked at grade 3 tumors and found that there was a benefit in adding immunotherapy to chemotherapy in high-risk, grade 3, early-stage HR-positive breast cancers. Based on [data from] our study, these trial findings may be driven by increased immune cell infiltration in high-grade, primary HR-positive tumors.”

Investigators also noted that a higher Oncotype DX score was similarly correlated with increased immune cell infiltration in HR-positive/HER2-negative breast cancer across all the immune cell infiltration markers examined. PIK3CA mutation status was not associated with a change in immune cell infiltration.

“We need to [determine] which patients with HR-positive primary breast cancer should be receiving immune therapy; if we can better characterize the immune cell populations in these tumors, it can hopefully [allow] us to avoid overtreating all patients with HR-positive primary breast cancer and instead focus on those who are most likely to benefit,” O’Meara said. “[In the future], we are interested in [evaluating] how the different immune cell populations in the tumors are correlated with response to both chemotherapy and immunotherapy.”

Refining Telehealth for the Delivery of Cancer Care

In her study, Natsuhara examined the impact of telehealth on patient diversity and care use at the UCSF Helen Diller Family Comprehensive Cancer Center from 2017 to 2019 (pretelehealth cohort) and 2021 to 2023 (posttelehealth cohort).2 Data from 2020 were excluded due to the irregularity of the COVID-19 pandemic to better isolate the effects of telehealth on outcomes. Using the UCSF clinical data warehouse, Natsuhara evaluated patient demographics, visit data, and treatment data related to the use of telehealth in patients with breast, gastrointestinal (GI), and genitourinary (GU) cancers. In 2023, patients in the respective groups were using remote visits at rates of 26.6%, 70.3%, and 91.9%.

“I have done most of my medical training during the COVID-19 pandemic,” Natsuhara said in an interview with Oncology Fellows. “[During this time], we saw a rapid transition to telehealth in medicine and what stuck with me was despite the fact that we are now 5 years post-pandemic, we are still seeing a lot of video visits. Telehealth is embedded in the medical system in a way that it wasn’t before, and during training, I started to wonder whether it was improving access to care. We have seen that telehealth can be a great tool for patients who live in rural areas, those who don’t have access to big cancer centers, and patients who can’t afford to take time off work. At the same time, we all recognize that there are a lot of barriers to [telehealth] use, [such as] having an internet-enabled device, a quiet place to have a visit, and the digital [literacy] to [use it].”

During the pretelehealth period, 5942 unique patients with breast cancer were seen across 32,450 outpatient encounters. The median age at first visit was 58 years (IQR, 48-67). In the posttelehealth period, 11,944 patients were seen across 38,680 unique encounters and the median age at first visit was also 58 years (IQR, 48-68).

Patients with GI cancers seen during the pretelehealth (n = 8332) and posttelehealth (n = 13,301) periods had a median age of 61 years (IQR, 52-70) and 63 years (IQR, 53-71), respectively. The total number of outpatient encounters was 30,899 and 39,671, respectively.

In the GU cancers group, 10,849 unique patients were seen during the pretelehealth period compared with 22,569 in the posttelehealth period. The numbers of outpatient encounters during these respective periods were 39,167 and 60,958, and the median ages at first visit were 68 years (IQR, 61-73) and 70 years (IQR, 63-76), respectively.

“We [also] found that the age of patients who were using phone visits in the posttelehealth period was higher than [that of] the patients who are being seen either in person or by video,” Natsuhara said. “This [may] highlight the challenges that older adults sometimes have with using virtual modes, and particularly video; those patients may be opting to do more phone visits. This is interesting because there are data in the literature that [show] the quality of care and the patient relationship conducted over phone are not of the same caliber and quality that we see for video visits. It is important for patients to be face-to-face with their provider and have those nonverbal cues.”

In terms of patient demographics, patients with breast cancer in the pretelehealth and posttelehealth groups were from San Francisco (24.8% vs 32.2%, respectively), East Bay (17.3% vs 16.3%), North Bay (13.1% vs 12.8%), South Bay (12.8% vs 14.3%), and outside the Bay Area (32.1% vs 24.2%). In the pretelehealth and posttelehealth groups of patients with GI cancers, these respective rates were 24.1% vs 23.1%, 18.5% vs 18.7%, 12.9% vs 13.6%, 10.8% vs 12.0%, and 33.6% vs 32.5%. These respective rates during the pretelehealth and posttelehealth groups with GU cancers were 14.3% vs 17.8%, 16.5% vs 17.3%, 15% vs 15%, 10.6% vs 11.2%, and 44.7% vs 38.7%.

Although nearly all the percentages decreased from the pretelehealth to posttelehealth periods, the total number of patients seen via telehealth increased across all location demographics during the time frame; this increase was statistically significant in patients with breast and GU cancers from outside the Bay Area, Natsuhara noted.

In-person visits decreased from the pretelehealth to posttelehealth periods across the breast cancer (98.4% vs 71.5%, respectively), GI cancer (98.1% vs 43.1%), and GU cancer groups (89.4% vs 14.7%). Conversely, video visits rose across the respective groups, from 1.6% to 28.1%, 1.9% to 55.2%, and 3.1% to 81.4%.

“It’s not necessarily 1 cancer division that’s driving telehealth use, but it seems to be somewhat disease specific, and [related to] a culture within that clinic of how many visits are conducted virtually vs in person,” Natsuhara noted.

Additional findings by oncology department type showed that medical (2.1% vs 25%, respectively), surgical (0% vs 35.2%), and radiation oncology (0.2% vs 39.1%) visits via telehealth increased in patients with breast cancer in the pretelehealth and posttelehealth periods. Similarly, in the GI cancer arm these respective rates were 2.1% vs 54.8%, 1.4% vs 68.4%, and 0.3% vs 54.7%. In the GU cancer arm, the rates were 2.9% vs 86.8%, 23.6% vs 91.3%, and 0.8% vs 69.2%, respectively.

“[At UCSF] we have the benefit of having our telehealth program rolled out in 2015, prior to the [COVID-19] pandemic, so we had some historical data to [help us] understand what this increase in telehealth had done to our patient diversity, as well as the numbers and types of patients who are accessing [telehealth] care at our center,” Natsuhara said. “An [important remaining] question I have is, how can we use telehealth to improve access and diversity in our clinical trial patient populations? [In the future], we’ll be examining clinical trial enrollment before and posttelehealth and running additional regression analyses to better understand the nuances and treatment factors [that] matter for patients in using video visits [to] access care.”

References

1. O’Meara T. ImmunoPROFILE: profiling tumor-associated immune cell populations in primary and metastatic breast cancer tissue. Presented at: OncLive SABCS Fellows Forum; December 9, 2024; San Antonio, TX.
2. Natsuhara KH, Rugo HS, Piawah S, et al. Telehealth and cancer care delivery: evaluating the impact of telehealth on patient access and treatment utilization at a comprehensive cancer center using real world data. Presented at: OncLive SABCS Fellows Forum; December 9, 2024; San Antonio, TX.