The identification of chromosomal rearrangements that result in oncogenic gene fusions ushered in the era of molecularly targeted therapies in oncology.
The identification of chromosomal rearrangements that result in oncogenic gene fusions ushered in the era of molecularly targeted therapies in oncology, beginning with the groundbreaking approval of imatinib (Gleevec) in 2001. This “magic bullet” targets the BCR-ABL1 gene fusion protein in patients with chronic myeloid leukemia.1
Thanks to the development of advanced sequencing technologies, a multitude of cancer-driving gene fusions have since been identified, prompting the development of a growing number of novel therapeutic options directed at such alterations. These agents include additional FDA-approved drugs that inhibit BCR-ABL protein activity as well as therapies aimed at ALK, RET, NTRK, and FGFR2 fusions.1,2
Investigators are seeking to add fusions involving the NRG1 gene to the list of targets. First described in 2014 in lung adenocarcinoma, the most prevalent subtype of non–small cell lung cancer (NSCLC),3 NRG1 fusions have now been identified at low rates across many cancer types (Figure 15).4,5 Moreover, a growing body of evidence demonstrates that these fusions are oncogenic and potentially targetable in malignancies in desperate need of novel treatment options.6,7
Figure 1. NRG1 Fusions by Tumor Type5
NRG1 fusions activate the HER2-HER3 tyrosine kinase receptor pair, making it a potential target of interest in cancers associated with that pathway. The outcomes of preclinical studies have suggested that repurposing FDA-approved HER inhibitors, such as afatinib (Gilotrif), may be a highly attractive therapeutic strategy for NRG1 fusion–positive tumors. A growing number of case reports show clinical efficacy of afatinib in such tumors or suggest potential for investigational drugs targeting HER3.8-15
Prospective clinical trials, including multicenter basket studies, are under way for several of these drugs (Table). If successful, these studies could result in therapies directed at NRG1 fusions joining the growing global pipeline of tumor-agnostic therapies.
Table. Ongoing Clinical Trials in NRG1 Fusion–Positive Tumors
The HER family (also called ERBB) is a group of 4 closely related cell-surface receptors that, when activated by their cognate ligands, trigger intracellular signaling pathways orchestrating vital cellular processes. Dysregulation of many of these processes leads to hallmarks of cancer.16,17
The NRG1 (or HRG1) gene encodes the neuregulin-1 (or heregulin) protein, the best-characterized member of a family of proteins that function as ligands for HER3 and HER4. NRG1 can be expressed as a number of different isoforms, all of which possess an epidermal growth factor (EGF)–like signaling domain through which they activate their receptors.6,7,18
Most NRG1 isoforms are transmembrane proteins, some of which undergo proteolytic processing to release a soluble version of the ligand. Thus, NRG1 can activate HER receptors on the same cell in which it is expressed (via autocrine signaling) and on neighboring or more distant cells (via paracrine signaling).6,7,18
Binding of their ligands triggers a conformational change in EGFR (also known as HER1), HER3, and HER4, exposing their dimerization domain. This allows these HER receptors to form a pair with a second HER receptor of either the same (homodimerization) or a different kind (heterodimerization). Dimerization triggers the intrinsic kinase activity of the receptor and initiates intracellular signal-ing cascades (Figure 26).6,7,18
Figure 2. The Role of NRG1 Fusions in Signaling Networks6
Both HER2 and HER3 rely on heterodimerization for their activation, HER2 because it has no known activating ligand and HER3 because it has little or no intrinsic kinase activity compared with the other HER family members.16,17
HER2 is the preferred dimerization partner for HER3,19 which contains multiple binding sites for the p85 regulatory subunit of the PI3K protein. Thus, HER2-HER3 heterodimerization is a key amplifier of PI3K/AKT signaling, in addition to a trigger for other signaling pathways, such as the MAPK pathway.16,17
In 2014, investigators identified a new type of genomic aberration in lung adenocarcinomas, in which the first 6 exons of the CD74 gene (containing its transmembrane domain, among other features20) were fused to the exons of the NRG1 gene that encode its EGF-like receptor-binding domain.
The investigators went on to demonstrate that this gene fusion drives cancer development because the NRG1 EGF-like domain in the resulting fusion protein is overexpressed on the cell surface; this type of cancer is otherwise devoid of NRG1 expression. The NRG1 EGF-like domain functions as a ligand for HER3 and triggers HER2-HER3 heterodimerization and subsequent intracellular signaling via the PI3K/AKT pathway.2,21
Up to that point, known drivers of lung cancer primarily involved alterations in tyrosine kinase receptors or downstream signaling kinases. NRG1 fusions represented a unique mechanism of oncogenesis, whereby signaling through a tyrosine kinase receptor was affected indirectly through aberrant expression of its ligand.
Since that initial report, NRG1 fusions have been identified across a growing number of cancer types. To date, the largest analysis retrospectively examined more than 40,000 tumor specimens using comprehensive genomic profiling.4
Across all tested samples, the incidence of NRG1 fusions was 0.2, but this varied among tumor types. More than half the fusions were identified in patients with NSCLC; however, just 0.3% of tested NSCLC specimens were fusion positive. NRG1 fusions were also identified in gallbladder, pancreatic, kidney, ovarian, breast, bladder, and colorectal cancers, as well as sarcomas.4,5
A range of NRG1 fusion partners beyond CD74 have been identified, and they also vary among cancer types. In NSCLC, other fusion partners include the SDC4, SLC3A2, and ATP1B1 genes. In other malignancies, the genes SETD4, TSHZ2, and ZMYM2 (ovarian cancer); ADAM9 and COX10-AS1 genes (breast cancer); and CDH1 and VTCN1 genes (pancreatic cancer) are among known fusion partners of NRG1.5
Most NRG1 fusions described to date result in the formation of a chimeric protein that like CD74-NRG1, possesses an intact EGF-like domain from NRG1 and a transmembrane domain donated by the fusion partner.11,22
NRG1 fusions have been shown to be associated predominantly with tumors with adenocarcinoma histology. In lung cancer, they appear to be particularly enriched in the invasive mucinous adenocarcinoma (IMA) subtype, a rare form of the malignancy that is associated with poor prognosis and rarely demonstrates currently targetable driver alterations, such as EGFR mutations or ALK fusions.5,23
NRG1 fusions have largely been shown to be mutually exclusive with other oncogenic drivers across cancer types, although exceptions have been reported. Notably, in pancreatic ductal adenocarcinoma, NRG1 fusions are found predominantly in KRAS wild-type tumors and could offer a much-needed target for precision medicine in this notoriously therapeutically challenging tumor type.5,11,12
The rarity of NRG1 fusions and the diversity of fusion partners may make detecting these fusions more challenging. Historically, fluorescence in situ hybridization and reverse transcriptase–polymerase chain reaction have been used to detect gene fusions, but these have several inherent disadvantages.6,7,18,24
DNA sequencing–based assays can be used to detect gene fusions, but the large intronic regions involved in NRG1 fusions can present an impediment, and current assays cannot detect all NRG1 fusions. RNA sequencing is now considered the gold standard for gene fusion detection because, compared with DNA-based methods, it has higher sensitivity for genomic rearrangements. It offers the greatest likelihood of identifying NRG1 gene fusions when the identity of the fusion partner is unknown.6,7,18,24
Because NRG1 fusions drive cancer through HER receptors and downstream signaling pathways, HER-targeted agents make attractive therapies for patients whose tumors harbor these fusions. Preclinical studies have demonstrated activity in NRG1 fusion-positive cancer cell lines and durable tumor regression in patient-derived xenograft models for a number of HER family-targeted drugs.
Furthermore, case reports and small retrospective studies demonstrating clinical activity of HER-targeted drugs in patients with NRG1 fusion–positive tumors are generating even greater enthusiasm and have propelled several drugs into ongoing prospective trials.
The largest body of evidence exists for afatinib, a pan-HER inhibitor that is already approved for the treatment of patients with EGFR-mutated NSCLC. Although EGFR inhibition was relatively ineffective at suppressing growth of NRG1 fusion–driven cancer in cells,8 afatinib may be more successful because it targets other HER family members in addition to EGFR and thus blocks the major mechanism of NRG1-driven oncogenesis, which is through HER2/ HER3 signaling.
A number of published case reports describe patients with NRG1 fusion–positive lung adenocarcinoma who responded to afatinib.9,10,14 Responses have also been noted in patients with other types of solid tumors, including cholangiocarcinoma and pancreatic cancer.10-12
Results for case reports published to date, in addition to 4 previously unreported cases, were summarized in a presentation at the Breakthrough: A Global Summit for Oncology Innovators, which the American Society of Clinical Oncology sponsored in October 2019. Afatinib was used to treat 17 patients with NRG1 fusion–positive advanced solid tumors (11 patients with lung adenocarcinoma; 3, pancreatic adenocarcinoma; and 1 each, cholangiocarcinoma, ovarian cancer, and colorectal cancer). Eleven patients experienced a partial response (PR), and 4 achieved stable disease (SD).15
Clinical activity of combined inhibition of EGFR and HER2 with erlotinib (Tarceva) plus pertuzumab (Perjeta) was also noted in a patient with NRG1 fusion–positive pancreatic adenocarcinoma.12 Preclinical activity also was observed with tarloxotinib, a dual EGFR/HER2 inhibitor activated by hypoxia, in NRG1-altered cell lines and NRG1 fusion–positive patient-derived xenograft models.25
Because NRG1 is an activating ligand for HER3, NRG1 fusions could represent an alternative mechanism of HER3 pathway dysregulation in tumors that do not overexpress the receptor, which suggests a rationale for testing HER3-targeted antibodies in patients with these gene fusions.
Increased levels of HER3 expression have been noted across numerous cancer types and are associated with tumor development and resistance to a variety of anticancer therapies. For this reason, ongoing efforts to target HER3 have met with limited success, generally owing to insufficient efficacy.16,17,26
Merrimack Pharmaceuticals developed the HER3-targeted monoclonal antibody seribantumab (MM-121), and it was tested in randomized phase 2 clinical trials in NSCLC and ovarian and breast cancers.27-29 Seribantumab did not demonstrate significant clinical benefit in unselected patient populations, but the results of biomarker analyses suggested that a subset of patients with NRG1-expressing tumors derived benefit.27-29
However, in the phase 2 SHERLOC trial (NCT02387216), adding seribantumab to docetaxel in patients with previously treated NRG1-positive NSCLC failed to improve progression-free survival compared with docetaxel alone. NRG1 expression was assessed using immunohistochemistry (positive score, ≥ 1 +). Among 109 patients treated and available for assessment, the median progression-free survival for participants who received the seribantumab combination (n = 71) was 3.4 months (95% CI, 2.6-4.2) versus 4.1 months (95% CI, 2.8-6.3) for those who took docetaxel alone (n = 38), translating into an HR of 1.382 (95% CI, 0.813-2.350; P = .2302). The trial was subsequently terminated.30
In 2019, Elevation Oncology acquired seribantumab31 and is currently carrying out the phase 2 CRESTONE trial in patients with NRG1 gene fusions (NCT04383210). The study, which aims to enroll 75 patients, will identify NRG1 fusions through molecular assays on fresh or archived formalin-fixed paraffin-embedded tumor samples.
Meanwhile, several case reports suggest that HER3-targeted antibodies may have activity in patients with NRG1 fusions.
One patient with previously treated CD74-NRG1 fusion–positive IMA achieved a PR lasting more than 18 months when receiving GSK2849330, an anti-HER3 antibody, during a phase 1 study (NCT01966445).8 Two patients with IMA harboring SLC3A2-NRG1 fusions were treated with erlotinib plus a different HER3-targeted antibody, lumretuzumab (RO5479599), and both achieved SD lasting nearly 4 months (NCT01482377).32
Dual targeting of both HER2 and HER3 may be more effective than blocking HER3 alone. Zenocutuzumab (MCLA-128) is a novel HER2/HER3–bispecific antibody described by developer Merus as having a dual “dock and block” mechanism of action. It binds to HER2 at a different epitope from that targeted by trastuzumab (Herceptin) and blocks the interactions between HER3 and both NRG1 and HER2.33,34
Zenocutuzumab was recently awarded orphan drug designation for the treatment of pancreatic cancer following signs of clinical activity in patients with this cancer type and NSCLC. In trial results reported at the 2019 AACR-NCI-EORTC Molecular Targets and Cancer Therapeutics international conference, 29 patients with NRG1 fusion–positive tumors across 8 cancer types were treated with zenocutuzumab at a dose of 750 mg every 2 weeks.
One patient with pancreatic adenocarcinoma with liver metastases achieved a PR, described as a reduction of cancer antigen (CA) 19-9 from 262 U/mL to 56 U/mL and a 44% reduction in tumor diameter. A second patient with pancreatic ductal adenocarcinoma with liver metastases achieved a reduction in CA 19-9 from 418 U/mL to 11 U/mL and a 22% reduction in tumor diameter. A patient with NSCLC and brain metastases who had progressed after 6 prior lines of therapy, including afatinib, experienced a PR, reported as a 33% reduction in tumor diameter and shrinkage of brain metastases.13 Zenocutuzumab is also being evaluated in the phase 1/2 eNRGy trial in patients with advanced NRG1 fusion–positive solid tumors (NCT02912949).