**Use of HER2-Directed Therapy in Metastatic Breast Cancer and How Community Physicians Collaborate to Improve Care**

**Joanne E. Mortimer 1,\*, Laura Kruper 2, Mary Cianfrocca 1, Sayeh Lavasani 1, Sariah Liu 1, Niki Tank-Patel 1, Mina Sedrak 1, Wade Smith 1, Daphne Stewart 1, James Waisman 1, Christina Yeon 1, Tina Wang <sup>1</sup> and Yuan Yuan <sup>1</sup>**


Received: 6 May 2020; Accepted: 18 June 2020; Published: 24 June 2020

**Abstract:** The development of new HER2-directed therapies has resulted in a significant prolongation of survival for women with metastatic HER2-positive breast cancer. Discoveries in the laboratory inform clinical trials which are the basis for improving the standard of care and are also the backbone for quality improvement. Clinical trials can be completed more rapidly by expanding trial enrollment to community sites. In this article we review some of the challenges in treating metastatic breast cancer with HER2-directed therapies and our strategies for incorporating our community partners into the research network.

**Keywords:** breast cancer; community; research; HER2-directed therapy

#### **1. Introduction**

Although breast cancers arise from a single organ, the biology and natural history of the disease can be extremely variable. Gene expression profiling allows us to subcategorize breast cancer into four "intrinsic subtypes", each with a unique natural history and response to therapy—Luminal A, Luminal B, Basal-like, and HER2 (human epidermal growth factor receptor 2) "enriched" [1,2]. However, until uniform genomic profiling becomes feasible, clinical decision making is based on tumor histology, stage, hormone receptor status, and HER2 status. This paper focuses on the use of HER2-targeted therapies in metastatic breast cancer and the importance of clinical trials and community practice collaboration in understanding the biology of this disease and advancing therapy.

#### **2. HER2 and Breast Cancer**

Four membrane tyrosine kinase receptors constitute the HER family. HER1 (also known as EGFR; epidermal growth factor receptor), HER3, and HER4 bind to almost a dozen different ligands, while HER2 has no known ligands. Rather, HER2 is activated through homo- and heterodimerization with other HER family members, and these complexes activate intracellular signaling pathways such as MAPK (mitogen-activated protein kinase) and PI3K (phosphoinositide 3-kinase) that results in tumor growth, invasion, migration, and survival [3]. The different receptors, ligands, and intracellular signaling pathways provide opportunities for drug development with agents used alone, in combination with each other, and with chemotherapy. The gene that encodes HER2 is amplified in 15–20% of

all breast cancers and defines a uniquely aggressive natural history with high grade histologies, greater likelihood to metastasize to visceral sites, and shortened survivals [4,5]. Gene amplification or protein over expression identifies patients who are candidates for HER2-directed therapies.

The humanized IgG monoclonal antibody, trastuzumab, targets the extracellular domain of HER2 effectively preventing intracellular signaling. While trastuzumab has minimal activity as a single agent, when combined with chemotherapy there is a prolongation in survival for women with metastatic breast cancer [6–8]. Trastuzumab has been shown to improve disease outcomes in all stages of HER2-amplified breast cancer [9]. The development of HER2-directed therapies, and trastuzumab in particular, is one of the greatest success stories in the treatment of breast cancer and is as important as hormone receptor status in its clinical impact.

#### **3. Assessment of HER2 Status**

Candidacy for HER2-directed therapies is based on a pathologic determination from the primary tumor or metastatic focus of disease. It is generally agreed that HER2 positivity is defined as 3+ staining by immunohistochemical staining (IHC) of the protein or gene amplification of HER2 by fluorescence in situ hybridization (FISH). Recently, companion studies described using in situ hybridization to report gene copy number per cell. This resulted in a HER2 "equivocal" status and required modification of the ASCO/CAP (American Society of Clinical Oncology/College of American Pathologists) guidelines for HER2 assessment [10].

The standard of care dictates that a tumor biopsy be performed on all patients suspected of having recurrent or metastatic breast cancer in order to document the disease as well as to determine the hormone receptor and HER2 status, which guide the choice of systemic therapy. The estrogen receptor (ER), progesterone receptor (PR), or HER2 status on a recurrence may differ from the original primary in 20–30% of instances. Furthermore, although it is not feasible to biopsy all sites of metastatic disease, discordance of receptors within an individual patient occurs in 15–20% of lesions [11–13]. Ultimately, the choice of systemic therapy is determined on a small sample of tumor obtained on a larger site of disease and thus may not reflect the tumor status at all sites.

HER2-directed therapies are effective only in patients whose cancers are HER2-positive. The NSABP B47 randomized over 3000 women with early stage breast cancer whose HER2 status was negative (1+, and 2+ by IHC of FISH< 2.0) to receive conventional adjuvant chemotherapy with or without 12 months of trastuzumab. No benefit was seen with the addition of trastuzumab and cardiac toxicity was observed in 2.3% receiving trastuzumab [14]. It is probably not realistic to assume that all HER2-positive patients benefit from HER2-directed therapies. Given a predictable incidence of cardiac toxicity, it would be ideal to treat only women who are likely to benefit from these treatments. The goal of precision medicine is to identify which patients benefit from HER2-directed therapies.

#### **4. Functional Imaging to Predict Response to HER2 Therapies**

We have developed 64Cu-labeled trastuzumab as a PET imaging agent to study women with recurrent/metastatic breast cancer. In our experience uptake on 64Cu-trastuzumab PET/CT correlates with the qualitative assessment of HER2 by IHC. Higher uptake of 64Cu-trastuzumab is seen in women with IHC levels of 3+ compared to 1+ [15,16]. Functional imaging with 64Cu-trastuzumab PET/CT does not assess HER2 status, but demonstrates that trastuzumab is effectively delivered to the cancer. This pharmacodynamics data is clinically more relevant than knowing the intensity of uptake by IHC or degree of gene amplification. The current standard of care for patients with newly diagnosed metastatic HER2-positive cancer utilizes a combination of chemotherapy, trastuzumab, and pertuzumab. It is, therefore, not possible for functional imaging with radiolabeled trastuzumab to predict impact of trastuzumab on treatment efficacy when response to therapy is confounded by the concurrent administration of chemotherapy.

Ado-trastuzumab emtansine (TDM1) binds to the extracellular domain of HER2 and undergoes receptor-mediated internalization inhibiting intracellular signaling pathways [17]. Because its activity is dependent upon trastuzumab binding to the HER2 protein, it is the ideal drug to test whether functional imaging with 64Cu-trastuzumab PET/CT is able to predict for response to therapy. In our experience, pretreatment 64Cu-trastuzumab PET/CT in women with recurrent/metastatic disease is able to identify which individuals are unlikely to benefit from TDM1. Additionally, 64Cu-trastuzumab PET/CT has identified disease within the CNS and has also demonstrated that uptake of 64Cu-trastuzumab can be heterogeneous within a single patient. If that uptake in individual sites falls below a certain level, TDM1 is not effective in those areas and the response is mixed [18].

#### **5. Current Treatment for Metastatic HER2-Positive Breast Cancer**

Like trastuzumab, pertuzumab is also a monoclonal antibody that binds to the extracellular domain of HER2, but at a different epitope from that bound by trastuzumab. Pertuzumab prevents dimerization with other HER family members (HER1, HER3, and HER4) [19]. Neither trastuzumab or pertuzumab are very active as single agents, but the combination is synergistic. In women with metastatic HER2-positive breast cancer that has progressed on a trastuzumab-containing regimen, the combination of pertuzumab with trastuzumab produces clinical benefit in more than half of patients [20]. The CLEOPATRA study demonstrated that the addition of pertuzumab to trastuzumab and docetaxel improved overall survival and established this three-drug combination as first-line therapy in metastatic disease. With more than 8 years of follow up, 37% of women who started treatment with pertuzumab, trastuzumab, and docetaxel are still alive. The study was designed to administer at least 6 cycles of docetaxel or until toxicity, and the duration of docetaxel was left to the discretion of the treating oncologist. Pertuzumab and trastuzumab were continued until disease progression. The median number of treatment cycles containing docetaxel was 8 (range 6–10) and the median total cycles was 24 with a maximum of 167 [21,22]. The prolonged survival in this population is also related to the efficacy of subsequent treatment.

Two antibody drug conjugates (ADC) have utilized trastuzumab to deliver a cytotoxic payload. Ado-trastuzumab emtansine (TDM1) utilizes maytansine as the cytotoxic component and has been found to be superior to other regimens containing chemotherapy and HER2-directed agents. It is currently considered second or third line therapy [17,23]. Fam-trastuzumab-deruxtecan uses the topoisomerase I inhibitor, deruxtecan as its payload. In the DESTINY 01 trial, 184 women with HER2-positive metastatic breast cancer who had received a median of six prior therapies were treated with fam-trastuzumab-deruxtecan. Objective responses were observed in 112 (60.9%) women with a median duration of response of 14.8 months [24]. Fam-trastuzumab-deruxtecan is used as third-line therapy.

The two ADCs differ in other ways. Fam-trastuzumab-deruxtecan has a higher drug-to-antibody ratio than TDM1 (DAR = 8 vs. DAR = 3.4, respectively) and fam-trastuzumab-deruxtecan is selectively cleaved by cathepsins that are up regulated in tumor cells. Deruxtecan is also highly cell-membrane permeable and is readily released across cell membranes. A potential benefit for this mechanism is that release of the cytotoxic component may have cytotoxic effects on adjacent tumor cells regardless of the tumor's HER2 dependency [25]. Since the majority of breast cancers (~60%) express HER2 to some degree, fam-trastuzumab-deruxtecan has the potential to benefit many patients [26,27]. Currently the drug is considered for 3rd line and beyond [28].

The oral tyrosine kinase inhibitor, tucatinib, is a selective inhibitor of HER2. In combination with capecitabine and trastuzumab, objective tumor responses have been demonstrated in heavily pretreated patients with metastatic HER2-positive breast cancer including women with brain metastases. The HER2CLIMB study randomized women with metastatic HER2-positive breast cancer who had received prior trastuzumab, pertuzumab, and TDM1 to receive trastuzumab and capecitabine with or without tucatinib. Progression-free survival (PFS) favored the addition of tucatinib (7.8 months compared with 5.4 months). The overall survival also favored the tucatinib arm with a survival prolongation of 4.5 months; 21.9 months vs. 17.4 months. For the 291 women with brain metastases, the median PFS was 7.6 months in the tucatinib arm and 5.4 months in the control arm (*p* < 0.001) [29]. Tucatinib was recently approved by the FDA. Tucatinib + trastuzumab + capecitabine is currently considered as a treatment for advanced unresectable or metastatic HER2-positive breast cancer, including patients with brain metastases, who have received one or more lines of prior HER2-targeted therapy in the metastatic setting [30]. In the HER2CLIMB study, 291 patients were enrolled who had brain metastases—48% in the tucatinib arm and 46% in the control arm. The CNS (central nervous system) progression-free survival and median overall survival favored the tucatinib arm making this an important therapy in patients with HER2+ breast cancer metastatic to brain [31].

There are additional active agents that also deserve mention. Lapatinib was the first dual inhibitor of HER2 and HER1. Until the development of TDM1, lapatinib and capecitabine were considered to be second-line therapy [32]. Currently, lapatinib is combined with trastuzumab as a "next-line" of therapy [33]. The NALA trial randomized patients who had received at least two prior treatments for metastatic disease to receive either neratinib or lapatinib in combination with capecitabine. At 6 and 12 months, the progression-free survival was significantly longer with neratinib as well as a significant increase in the time to intervention for symptomatic CNS. Neratinib provides another treatment option for treating metastatic disease [34]. Even after disease progression on trastuzumab, additional chemotherapy with trastuzumab has been shown to be more effective than chemotherapy alone [35].

The explosion in new and highly effective therapies has transformed metastatic HER2-positive breast cancer into a chronic disease. Conventional chemotherapy still has a role as salvage treatment and is more effective when combined with trastuzumab [35]. New drug development and combining the currently available HER2-directed agents with drugs that modulate intracellular signaling are currently on-going in the clinic.

#### **6. HER2 Activating Mutations—A Unique Clinical Entity**

Bose used DNA sequencing on samples obtained from ACOSOG (American College of Surgeons Oncology Groupz) 1031 and data from other sequencing studies to identify somatic mutations in tumors that were pathologically determined to be HER2-negative. Seven of the 13 mutations were classified as activating mutations. Cell line studies showed a unique dependence on EGFR phosphorylation and led to an assessment of the therapeutic efficacy of the dual kinase inhibitors of HER2 and EGFR, lapatinib and neratinib. Neratinib demonstrated a stronger inhibition of cell growth than lapatinib and was effective in all activating mutations [36]. It is estimated that HER2-activating mutations occur in 1.6–2.5% of invasive ductal carcinomas and 7.5% of invasive lobular cancers [36,37]. Clinical trials of neratinib in this population have demonstrated efficacy and have identified another way of targeting HER2 [38,39]. This unique genotype demonstrates the potential use of next generation sequencing in breast cancer.

#### **7. Engaging the Community in Clinical Research**

City of Hope recently acquired 30 community practice sites and in doing so dramatically expanded the number of colleagues in surgical, medical, and radiation oncology and almost tripled the number of new breast patients. With such a rapid expansion, our goal was to develop a common culture of research and quality clinical care that is at the heart of City of Hope's values. This initiative was led by a steering committee of interdisciplinary subspecialty physicians and business leaders who met regularly. This team identified the individual community physicians who had an interest in treating breast cancer and in conducting clinical research and worked with all the stakeholders to define quality of care. We convened a meeting of all clinical faculty with an interest in breast cancer, compiled a brief resource inventory (systematic identification and access to any potential contributions to this program), and identified barriers to achieving our predefined quality of care. The sharing of our unique differences in practice and the challenges that each individual experienced, resulted in a respect for what each partner brought to the program and how we could work together to solve these problems. Through a culture of respect and trust we were able to define quality metrics for the clinical care of

patients with breast cancer and created treatment guidelines which included eligibility for clinical trial participation (see Table 1). Clinical research requires physicians who are interested in conducting clinical trials and a practice site that has a patient population interested in study participation. It is also critical to have competent clinical research coordinators, a research pharmacy, and radiologists who support serial radiologic interpretation of x-rays using RECIST and PERCIST.


**Table 1.** City of Hope Institutional guidelines HER2+ metastatic disease.

All research decision-making is made by the breast cancer research team (Breast Disease Team), composed of medical, surgical, and radiation oncologists, basic scientists, statisticians, pharmacists, research nurses, clinical research associates, and the Executive Director for Affiliate sites. Breast Disease Team meetings are available on a remote platform so that our community partners can participate, and meeting minutes are submitted to all. The Breast Disease Team meets on a weekly basis to review proposed study concepts, endorse the written protocols developed from an approved concept, troubleshoot operational problems, and review active patients on study. Our current clinical research portfolio in HER2-positive breast cancer focuses on two main areas: 1. Defining new therapies and new roles for these systemic therapies (an area largely driven by the cooperative groups and Pharma), and 2. Identifying which patients are likely to benefit from HER2-directed therapies using functional imaging, which is a City of Hope research initiative.

The Executive Director for Affiliates oversees research at the community sites and is aware of the physicians' interests, patient population, and clinical trial infrastructure at the individual offices. The full clinical trial portfolio is not activated at every site and phase I clinical trials are generally restricted to the main campus. The most successful clinical trials to conduct in the community are those therapeutic trials that focus on improving disease outcomes through the use of new agents or testing approved agents used in different settings. For example, mutations in PIK3A have been associated with worse outcomes in adjuvant therapy trials and in CLEOPATRA [21,40]. A phase III NRG trial builds on the positive results of CLEOPATRA in women with metastatic HER2-positive disease and tests the benefit of adding the immune checkpoint inhibitor, atezolizumab, as first line therapy. A Novartis sponsored trial randomizes women with metastatic disease and a PIK3A mutation to a placebo controlled double blind trial of alpelisib added during the patients' maintenance of trastuzumab and pertuzumab. We are currently enrolling patients with documented HER2-activating mutations to a phase II trial sponsored by PUMA of neratinib, trastuzumab, (and fulvestrant if also ER-positive).

Our radiolabeled trastuzumab imaging studies require a radiopharmacy and expertise by a committed team of radiologists and physicist who interpret the images. Our community oncologists play an important role in these studies by referring patients, thereby allowing us to complete accrual expeditiously. They are authors on our publications and are recognized by their local communities as true experts who participate in innovative research.

#### **8. Optimizing Partnerships**

The Breast Disease Team has a number of initiatives to ensure inclusion of our community partners in research, and the weekly Breast Disease Team meeting is at the heart of these collaborations. Discussions at weekly Tumor Boards and a bi-weekly treatment planning meeting frequently highlight on-going studies and identify study candidates. City of Hope is a founding member of the NCCN (national comprehensive cancer network) and has a long history of participating in guideline panels. While some of the guidelines for treating breast cancer are definitive, others offer acceptable treatment options. The breast cancer team on the main campus meets on an annual basis to compile our guideline preferences for treatment, including, not only standard of care but the open studies for clinical trial participation as well (see Table 1).

#### **9. Future Directions**

We continue to activate investigator-initiated, cooperative group, and Pharma trials in the community that focus on improving the outcomes for women with metastatic HER2-positive disease. Our imaging research will utilize radiolabeled trastuzumab to predict for response to new agents such as fam-trastuzumab-deruxtecan and tucatinib [24,29]. These studies will rely on our community partners to refer interested patients for study participation.

National Cancer Institute-designated Comprehensive Cancer Centers (NCI-CCC) such as City of Hope strive to provide expert interdisciplinary clinical care and cutting-edge treatments and technologies. Between 1998 and 2008, 6.4% of adults between 22 and 65 years of age who were diagnosed with cancer in Los Angeles County received treatment at an NCI-CCC. For all primary tumor sites, the survival of patients treated at an NCI-CCC was superior to patients treated at a non-NCI-CCC. In this study, 6.5% of the 31,764 breast cancer patients included in the analysis were treated at an NCI-CCC and their survival was superior to those treated at non NCI-CCC sites [41]. Realistically, the majority of cancer patients will receive their care in community sites, and it is an important part of the mission of a Comprehensive Cancer Center to ensure that the standard of care in the community is comparable to the care that is delivered in these large research centers. Clinical research is at the heart of improving quality care.

**Author Contributions:** Conceptualization and writing, J.E.M., L.K. and Y.Y.; writing, review and editing, M.C., S.L. (Sayeh Lavasani), S.L. (Sariah Liu), N.T.-P., M.S., W.S., D.S., J.W., C.Y. and T.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors thank Nicola Welch, CMPP for editorial assistance and critical review of the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **NCI–Clinical Trial Accrual in a Community Network A**ffi**liated with a Designated Cancer Center**

#### **Daniel J. Kim, Dan Otap, Nora Ruel, Naveen Gupta, Naveed Khan and Tanya Dor**ff **\***

City of Hope Comprehensive Cancer Center, Department of Medical Oncology and Developmental Therapeutics, Duarte, CA 91010, USA; danieljkim@coh.org (D.J.K.); dotap@coh.org (D.O.); nruel@coh.org (N.R.); ngupta@coh.org (N.G.); nakhan@coh.org (N.K.)

**\*** Correspondence: tdorff@coh.org

Received: 29 April 2020; Accepted: 17 June 2020; Published: 24 June 2020

**Abstract:** Most cancer care is delivered in the community, while most clinical trials exist in academic centers. We analyzed clinical trial accrual of a tertiary care cancer center and its affiliated community sites to better understand what types of trials accrued at the community sites and whether community accrual increased ethnic diversity. The institutional clinical trial database was searched for solid tumor accruals during 2018–2019. Patient's race was abstracted, and trial's funding source, phase, and disease type/stage were tabulated. Of 3689 accruals, 133 were at community sites, representing 26 unique trials while the main campus accrued to 93 unique trials. Community site accruals were highest for breast and colorectal cancer, but patients with less common cancers such as renal, nasopharyngeal, and gastric cancer were also accrued at community sites. Accruals occurred to randomized trials, as well as phase Ib and translational biomarker studies. Minority patients constituted 20.0% and 32.5% of community site accruals for therapeutic and non-therapeutic trials respectively, compared to 20.6% and 29.8% of main campus accruals for therapeutic and non-therapeutic trials, respectively. We conclude that community sites affiliated with an academic cancer center can accrue to a broad spectrum of clinical trials while enhancing racial diversity in participation of clinical trials. Further expansion of access to clinical trials in community sites is necessary to broaden patient access to state-of-the-art and next-generation treatment options.

**Keywords:** clinical trials; community practice; minorities; ethnicity; race

#### **1. Introduction**

Community oncology practices provide approximately 55% of all care for cancer patients in the United States [1]. Providing access to clinical trials in community oncology practices has been a major initiative of the National Cancer Institute (NCI, Rockville, MD, USA), as evidenced by the creation of the National Community Oncology Research Program (NCORP, Bethesda, MD, USA) in 2014. The NCORP was created by consolidating the previous NCI supported networks of the CCOP–community clinical oncology program and the NCI Community Cancer Centers Program (NCCCP, Bethesda, MD, USA). Up to a third of clinical trial accrual in NCI studies is contributed by these community oncology practices [2], yet the overall clinical trial accrual remains lower than desirable: fewer than 5% of adult cancer patients enroll in clinical trials [3], despite 70% of cancer patients expressing interest in enrolling in a clinical trial [4].

Aside from increasing treatment options and providing access to the latest innovations in cancer care, participation in clinical trials has been shown to improve overall treatment quality vis-à-vis utilization of optimal standards of care [5]. One barrier for community practice physicians has been cost; clinical trial enrollment takes time, which takes resources away from overall clinic flow, and research staff are required regardless of how robust accrual is. Understanding the types of clinical trials that readily accrue in the community setting could help improve trial selection to foster financial

sustainability, since fewer trials, each of which accrue larger numbers of patients, is more efficient in the setting of having limited clinical research staff at community sites.

In addition, inadequate representation of racially diverse cancer patients has been noted to be a major problem in the applicability of most clinical trials to everyday practice. For example, clinical trials in metastatic castration-resistant prostate cancer (CRPC) were found to have 80% under-representation of African American patients [6]. In the case of geriatric patients, a study of NCI-sponsored trials found that patients over 65 made up only 32% of participants, whereas they made up 61% of patients affected by the cancers [7]. Given that community oncology practices may be located closer to patients' homes, the availability of trials in a local setting could reduce potential barriers for patients to access clinical trials. Furthermore, if community practices are located near specific ethnic enclaves, their participation in clinical trials could help alleviate the lack of racial diversity, as indicated by the NCCCP initiative [8].

City of Hope is an NCI-designated comprehensive cancer center in Duarte, California, with an extensive network of 21 community practice sites located within its vicinity in Southern California, thus allowing for delivery of specialized care near where patients reside. As part of the overall mission of City of Hope, select clinical trials have been opened to accrual at six of the community sites which service a high enough patient volume to support research recruitment. We sought to characterize and evaluate clinical trials accrual with a focus on City of Hope's community sites, and whether racial diversity was enhanced in patients accrued at community sites compared to the main cancer center campus.

#### **2. Material and Methods**

After obtaining IRB exemption for the study, we identified the individual accrual events in our clinical trials database during the calendar years 2018 and 2019 at all City of Hope sites. This dataset was extracted from our institution's secure online trial database, devoid of any personal identifiers. Inclusion criteria was prospective adult clinical trials. Study title, sponsor, location of accrual, and date of study were extracted from the dataset. We decided ahead of time to exclude trials which were felt to reflect highly tertiary patient populations, such as hematologic malignancy trials which, at City of Hope, were likely to be heavily transplant and chimeric antigen receptor T cell therapy-based, or any other trials felt to introduce bias into the analysis.

Results were tabulated for comparison between main campus and community sites and stratified by the type of trials. The Chi-square and Fisher's exact tests were used to test the differences in distribution of race, minority status, and age groups between community sites and main campus participants.

#### **3. Results**

Of the 33,558 individual accrual events identified in our analysis, 23,809 accrual events were identified to be retrospective in nature and thus excluded, along with an additional 5625 accrual events which were identified to be trials involving hematologic malignancies. Sixteen pediatric trial accrual events were also excluded, as well as 419 healthy volunteer accruals. Also excluded were two high-accruing non-therapeutic trials specifically recruiting only Hispanic women, as ethnicity was not a variable and thus biasing the minority accrual question.

Our final analysis was thus based on 3689 accrual events. Community sites in the City of Hope network accrued to 26 distinct, solid tumor clinical trials (*n* = 22 therapeutic, *n* = 4 non-therapeutic) during the years 2018–2019, compared with accrual to 93 solid tumor trials at the main campus during the same time period. A description of all therapeutic trials which accrued at least one subject at a community site is delineated in Table 1. During the same time period, an additional 57 trials were open at community sites but did not accrue patients there.

**Table 1.** Characteristics of therapeutic clinical trial which accrued subjects at City of Hope community practice sites from 2018–2019.

Accrued at main campus, transitioned during study treatment to community site. GEJ = gastro-esophageal junction. IST = Investigator Sponsored Trials. NSCLC = non-small celllungcancer.

\* Overall, 133 subjects were accrued at 6 community sites (*n* = 58 therapeutic) and 3556 at the main campus. These included randomized phase II or III trials (*n* = 13), as well as (*n* = 7) phase II trials, and (*n* = 2) phase Ib trials. Half of the trials accrued to in the community sites (*n* = 13) were cooperative group trials, four were industry, one was NCI/CTEP sponsored, and seven were investigator-initiated trials commonly called investigator-sponsored trials (IST). Trials which did not register accrual at community sites were primarily phase II (30/57; 52.6%) and phase III (21/57; 36.8%), with only five phase I trials open and not accruing (8.8%).

Most trials that registered community accruals were for breast cancers (*n* = 8), followed by gastrointestinal (6), genitourinary (3), non-small cell lung cancer (3), melanoma (1), and all solid tumors (1). Highest accruing trials in the community sites included a phase III industry-sponsored trial in metastatic refractory colorectal cancer, which accrued more patients in the community than at the main cancer center (*n* = 10 compared to *n* = 4 respectively). The other top accruing studies (*n* = 6 in community) were for breast cancer patients, one for prevention and one for first-line treatment of metastatic disease. Studies which were open in the community, but did not accrue, were distributed across tumor types; 13 gastrointestinal (22.8%), breast and lung (each 19.3%), melanoma/sarcoma and genitourinary (each 14.0%), head and neck and gynecologic (each 5.3%).

The ethnic composition of accruals in the community sites and the main campus is summarized in Table 2. Minority accrual ranged from 20.6% to 21.8% at the main campus and 15.6% to 20.0% at community sites, although there was a higher rate of accrual of African American subjects to non-therapeutic trials at community sites. Distribution of patient age was different between accrual sites (Table 3) within therapeutic and non-therapeutic studies (*p* < 0.01), as we saw a larger proportion of older patients at community sites than on main campus. Among adult subjects, 22/58 (37.9%) subjects were > 70 years old in the community, compared to 498/1828 (27.3%) patients accrued in Duarte. In contrast, 238 (13.0%) adult subjects on main campus vs. zero patients at community sites were under the age of 40.




**Table 2.** *Cont*.

\* *p* values calculated for distribution of known race or ethnicity, omitting unknowns from calculations.

**Table 3.** Age distribution of subjects accrued to clinical trials at the main campus and at community sites.


A complete list of trials accrued to at each site can be found in Supplemental Table S1. Notably, trials accrued to at community sites included the full spectrum of disease, from cancer prevention to adjuvant and neoadjuvant trials, as well as trials for metastatic and refractory cancer patients. Non-interventional trials which accrued at community sites included biomarker studies, both for molecular characterization as well as for predicting toxicity, and geriatric oncology trials. Studies not open in the community included supportive care trials such as a trial studying bright light to reduce frailty during androgen deprivation therapy for prostate cancer patients and studies of the perception of genomic profiling and of immunotherapy expected outcomes. Most phase I trials were not open in the community (26/32).

#### **4. Discussions**

Community sites associated with an NCI-designated cancer center may represent a unique opportunity to expand clinical trial access into underserved and under-represented populations, given the proximity of the sites to serve their respective local communities and the ability to leverage research staff and resources from the main campus. A review of more than 33,000 individual accrual events at City of Hope revealed that affiliated community sites accrued patients to a wide selection of solid tumor trials of various cancer histology, stage of disease, and therapeutic/non-therapeutic aims. Whereas it was anticipated most trials that successfully accrued patients at community sites would be later phase, cooperative group trials, there were also a relatively high accrual of investigator-sponsored phase II, and even phase Ib, trials. Similarly, it was anticipated that the most common malignancies

would make up the largest percentage of accruals in the community sites, but uncommon tumor types such as hepatocellular, kidney, and pancreatic cancers also accrued at community sites.

City of Hope community practice sites are strategically located throughout five counties across Southern California (Ventura, Los Angeles, San Bernardino, Riverside, and Orange), aiming to provide cancer care to all communities, including socioeconomically disadvantaged populations and communities with cancer disparities. Our analysis did reveal a higher accrual of African American patients to non-therapeutic trials at the community sites compared to main campus but did not significantly improve diversity of trial accrual as had been hoped. Loree et al. [9] had reported African Americans and Hispanics representing 3.1% and 6.1% of trial participants respectively, and City of Hope's rate of African American representation in trial accrual was similarly low. A limitation of our analysis is that our percentages are derived from accrued subjects rather than all subjects; that is to say, the denominator are trial participants, rather than all screened patients or all eligible patients or all patients seen at the clinic, as such data was not available for analysis. Our assessment was also limited by a relatively high rate of "unknown" ethnicity among community accruals (22.4%); this can be due to incomplete datasets or patients declining to answer the question. In order to minimize missing data, standard operating procedures for research staff and centralization of clinical data management could improve consistency and data quality.

Community sites did enroll a higher rate of geriatric patients, which may reflect the difficulty elderly patients face in traveling greater distances to a comprehensive cancer center in order to obtain oncology care. Admittedly, the City of Hope experience may be unique, as geriatric oncology is a flagship program with dedicated resources. However, this finding is an encouraging signal, and warrants further study as increasing accrual of geriatric patients to cancer protocols is an important goal for the oncology community given that many cancers have substantially higher prevalence in people over the age of 65.

Additional limitation of our study involves the inherent selection process involved in extending clinical trial offerings to the community sites. Based on limited resources and anticipated accrual rates, only a select few clinical trials were extended from the main campus to the community sites, and of the 21 community sites only six higher-volume sites were selected for clinical trial participation. Trial accrual rate could have been higher for the community sites if trial offerings were identical across main campus and community sites, but such an offering would require far greater investment into trial resources at all of the community sites and may not be practical or feasible. Our analysis also could not capture the number of patients who were referred from the community sites to the main campus for enrollment into trials not offered at the community sites, thus shifting trial accrual numbers from community sites to main campus. We did notice, however, that a bladder cancer trial enrolled subjects at the main campus and then transitioned ongoing cycles of treatment to community sites due to patient travel burden (Table 1). This flexibility is an example of how academic-community collaboration can optimize both accrual and patient-centered care and could represent an efficient "just in time" model of opening the right trials in the right areas. We concur with the value of main campus regularly meeting with community sites to review currently available clinical trials [10], and we additionally propose that the community sites must be actively engaged in and collaborate with the main campus in selecting the trials offered at the community sites based on needs assessment of their unique patient population and community site clinician interests.

In contrast to published experiences, however, we found that phase I and even non-therapeutic trials did not present an insurmountable barrier to accrual in community oncology practices. In a previous survey of 51 community practices, which included both federally sponsored and private non-academic affiliated practices, half of the practices self-reported that they did not accrue to phase I trials, 53% did not accrue to investigator-initiated trials, and 33% did not enroll on correlative science trials [11]. It is likely that many investigator-initiated trials and translational "non-therapeutic" trials can meet the needs of community oncology patients, and that academic-affiliated community sites are best poised to engage patients in accrual to these types of studies.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2077-0383/9/6/1970/s1, Table S1: summary of phases and disease states of trials accrued to at community sites and City of Hope main campus.

**Author Contributions:** Conceptualization and methodology, T.D.; data curation, D.O.; validation, D.J.K., D.O., N.R., and T.D.; formal analysis, N.R. and T.D.; writing—original draft preparation, D.J.K., N.R., N.G., and N.K.; writing—review and editing, D.K., T.D.; project administration and funding acquisition, T.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by an internal core grant #P30CA033572 from City of Hope, and received no external funding.

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

#### *Article*
