NRG1 Gene Fusions Are Recurrent, Clinically Actionable Gene Rearrangements in KRAS Wild‐Type

Author Manuscript Published OnlineFirst on May 8, 2019; DOI: 10.1158/1078-0432.CCR-19-0191 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1 NRG1 gene fusions are recurrent, clinically actionable gene rearrangements in KRAS wild‐type

2 pancreatic ductal adenocarcinoma

4 Martin R. Jones1, Laura M. Williamson1, James T. Topham2, Michael K.C. Lee3, Angela Goytain4, Julie Ho4,

5 Robert E. Denroche5, GunHo Jang5, Erin Pleasance1, Yaoquing Shen1, Joanna M. Karasinska2, John P.

6 McGhie3, Sharlene Gill3, Howard J. Lim3, Malcom Moore6, HuiLi Wong3, Tony Ng4, Stephen Yip4, Wei

7 Zhang1, Sara Sadeghi1, Carolyn Reisle1, Andrew J. Mungall1, Karen L. Mungall1, Richard A. Moore1,

8 Yussanne Ma1, Jennifer J. Knox5,6, Steven Gallinger5,7, Janessa Laskin3*, Marco A. Marra1,8*, David F.

9 Schaeffer2,4*, Steven J.M. Jones1,8,9*, Daniel J. Renouf2,3*

11 1BC Cancer, Canada’s Michael Smith Genome Sciences Centre, Vancouver, BC, Canada; 2Pancreas Centre

12 BC, Vancouver, Canada; 3BC Cancer, Division of Medical Oncology, Vancouver, BC, Canada; 4Department

13 of Pathology & Laboratory Medicine Vancouver General Hospital, Vancouver, BC, Canada; 5PanCuRx

14 Translational Research Initiative, Ontario Institute for Cancer Research, Toronto, ON, Canada; 6Division

15 of Medical Oncology, Princess Margaret Cancer Centre, Toronto, ON, Canada; 7Lunenfeld Tanenbaum

16 Research Institute, Mount Sinai Hospital, Toronto, ON, Canada; 8Department of Medical Genetics,

17 University of British Columbia, Vancouver, BC, Canada; 9Department of Molecular Biology and

18 Biochemistry, Simon Fraser University, Vancouver, BC, Canada.

20 *Authors co‐supervised this work

22 Running title: NRG1 gene fusions in PDAC are clinically actionable

23 Keywords: PDAC, NRG1, fusion, actionable, recurrent

24 Additional information:

Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on May 8, 2019; DOI: 10.1158/1078-0432.CCR-19-0191 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

25 Financial support:

26 We acknowledge contributions towards equipment and infrastructure from Genome Canada and

27 Genome BC (projects 202SEQ, 212SEQ, 12002), Canada Foundation for Innovation (projects 20070,

28 30198, 30981, 33408) and the BC Knowledge Development Fund. This research was generously

29 supported through unrestricted philanthropic donations received through the BC Cancer Foundation, as

30 well as funding provided by the Terry Fox Research Institute, Pancreatic Cancer Canada and Genome

31 British Columbia (project B20POG). MAM acknowledges infrastructure investments from the Canada

32 Foundation for Innovation and the support of the Canada Research Chairs and CIHR Foundation (FDN‐

33 143288) programs.

35 Corresponding author:

36 Dr. Daniel Renouf, MD, MPH, FRCPC

37 Assistant Professor, University of British Columbia

38 Medical Oncologist, British Columbia Cancer Agency

39 600 West 10th Avenue, Vancouver BC V5Z 4E6

40 Tel: +1 (604) 877 6000 ext. 672445

41 Email: [email protected]

43 Stephen Yip has received consultation fees from Bayer and Pfizer, and travel allowance from

44 Roche/Foundation One. Janessa Laskin has received honorarium for academic talks from Roche, BI,

45 Astra Zeneca and Pfizer, and research funding from Roche, BI, Astra Zeneca and Eli Lilly. Daniel J. Renouf

46 received research funding/honoraria from Bayer, and honoraria from Servier, Celgene, Taiho and Ipsen.

48 Translational Relevance

49 Gene fusions involving Neuregulin 1 (NRG1) have been noted in multiple cancer types, and have

50 potential therapeutic implications. Here we describe three patients with advanced, KRAS wild‐type,

51 pancreatic ductal adenocarcinoma (PDAC) that were comprehensively profiled by whole genome and

52 transcriptome sequencing. All three tumours were positive for gene fusions involving the ERBB3 ligand,

53 NRG1. Two of the three patients were treated with the HER‐family kinase inhibitor afatinib and

54 demonstrated a significant and rapid response while on therapy. This work adds to a growing body of

55 evidence that NRG1 gene fusions are recurrent, therapeutically actionable, genomic events in

56 hepatobiliary/pancreatic cancers. Based on the clinical outcomes described here, patients with KRAS

57 wild‐type PDAC would benefit from routine testing for NRG1 gene fusions.

59 Abstract

60 Purpose:

61 Gene fusions involving Neuregulin 1 (NRG1) have been noted in multiple cancer types, and have

62 potential therapeutic implications. While varying results have been reported in other cancer types, the

63 efficacy of HER‐family kinase inhibitor afatinib in the treatment of NRG1 fusion positive pancreatic

64 ductal adenocarcinoma is not fully understood.

66 Experimental Design:

67 47 patients with pancreatic ductal adenocarcinoma received comprehensive whole genome and

68 transcriptome sequencing and analysis. Two patients with gene fusions involving NRG1 received afatinib

69 treatment, with response measured by pre‐ and post‐treatment PET/CT imaging.

71 Results:

72 3/47 (6%) of patients with advanced pancreatic ductal adenocarcinoma were identified as KRAS wild‐

73 type by whole genome sequencing. All KRAS wild‐type tumours were positive for gene fusions involving

74 the ERBB3 ligand NRG1. 2/3 patients with NRG1 fusion positive tumours were treated with afatinib and

75 demonstrated a significant and rapid response while on therapy.

77 Conclusions:

78 This work adds to a growing body of evidence that NRG1 gene fusions are recurrent, therapeutically

79 actionable, genomic events in pancreatic cancers. Based on the clinical outcomes described here,

80 patients with KRAS wild‐type tumours harbouring NRG1 gene fusions may benefit from treatment with

81 afatinib.

83 Introduction

84 Oncogenic gene fusions involving Neuregulin 1 (NRG1) are recurrent across a variety of different cancer

85 types and have emerged as clinically actionable genomic events in lung cancers and

86 cholangiocarcinoma(1–4). As part of the ongoing Personalized OncoGenomics (POG; NCT02155621) and

87 Prospectively Defining Metastatic Pancreatic Ductal Adenocarcinoma Subtypes by Comprehensive

88 Genomic Analysis (PanGen; NCT02869802) clinical trials at our site, patients with advanced cancers

89 undergo comprehensive genomic analysis including whole genome sequencing and transcriptome

90 analysis (WGTA). We have previously described a patient with cholangiocarcinoma that harboured an

91 ATP1B1‐NRG1 gene fusion who responded favorably to treatment with the HER‐family kinase inhibitor,

92 afatinib(2). However, not all tumours with NRG1‐fusions are responsive to afatinib, as evidenced by a

93 lack of therapeutic benefit reported in a series of NRG1‐fusion positive lung cancers(4), highlighting the

94 need for further research in this area.

96 At a genomic level, PDAC is defined by the presence of KRAS driver mutations in the vast majority of

97 cases, and are for the most part highly resistant to molecularly targeted therapeutic strategies. There is

98 however a subset of PDACs that are KRAS wild‐type. As part of the POG study, we have sequenced

99 tumour biopsies from 47 PDAC patients, 91% of which were characterized by clonal KRAS gain of

100 function mutations. KRAS wild‐type PDAC tumours are thought to be dependent upon alternate

101 oncogenic drivers that, though poorly understood(5), may provide opportunities for selective

102 intervention with appropriate targeted therapies. Here we present 3 cases of KRAS wild‐type PDAC that

103 harbour oncogenic NRG1 fusions, with two cases demonstrating a significant response when treated

104 with the HER‐family kinase inhibitor afatinib. NRG1 fusions were highly expressed, retained the

105 functional EGF‐like domain required for HER‐family kinase activation and involved transmembrane

106 domain‐containing fusion partners. A study published by Heining et al. similarly identified 3 NRG1‐fusion

107 positive cases in their cohort of 17 PDAC tumours, one of which also demonstrated a reduction in

108 tumour size in response to treatment with afatinib(6). Together, these observations provide evidence

109 for the recurrent nature of oncogenic NRG1 fusions in KRAS wild‐type PDAC and for the beneficial

110 treatment of afatinib in such cases, supporting further investigation of its use in this setting.

111

112 Materials and Methods

113 Patients were enrolled in the POG trial at BC Cancer in Vancouver, British Columbia. Patient biopsy

114 samples received whole genome and transcriptome profiling in order to identify potentially clinically

115 actionable genomic events. The study was approved by the University of British Columbia Research

116 Ethics Committee (REB# H12‐00137, H14‐00681) and was conducted in accordance with international

117 ethical guidelines. Written informed consent was obtained from each patient prior to genomic profiling.

118 Patient identity was anonymized and an identification code assigned to the case for communicating

119 clinically relevant information to physicians. Patients consented to potential publication of findings. Raw

120 sequence data and downstream analytics were maintained within a secure computing environment.

121

122 Whole‐genome and transcriptome analysis

123 Whole‐genome and transcriptome analysis (WGTA) was performed as previously described(2). Briefly, a

124 fresh tumor biopsy and blood sample were collected and sequenced to a mean redundant depth of

125 coverage of 40x and 80x, respectively and a transcriptome of approximately 200M reads generated from

126 the tumour sample. Somatic point mutations, small insertions and deletions (indels), and copy‐number

127 alterations, present in the tumor DNA but not in the germline were identified(7–9). Detection of

128 structural rearrangements was achieved by merging and annotating SVs detected by de novo assembly

129 (Trans‐ABySS)(10) and existing SV callers; deFUSE(11), DELLY(12), Chimerascan(13) and Manta(14); using

130 MAVIS(15). Fusion and RNAseq alignment visualization were generated using MAVIS. Publicly available

131 transcriptome sequencing data from The Cancer Genome Atlas (TCGA)(16) were used to evaluate the

132 expression profile of gene transcripts.

133

134 Mutation Signature Analysis

135 The mutation signature profile was determined by classifying all genomic SNVs into 96 classes based on

136 variant and 3’/5’ mutation context to obtain a mutation catalog vector as described by Alexandrov et

137 al.(17). In order to determine the best fit to a consensus set of 30 mutation signatures (available at

138 http://cancer.sanger.ac.uk/cosmic/signatures), we performed non‐negative least squares

139 decomposition implemented in the R package “nnls”.

140

141 Molecular Subtyping

142 Molecular subtyping was performed based on gene expression signatures detailed in Moffit et al.,

143 2015(18), Bailey et al. 2016(19), and Collisson et al., 2011(20). Given the relatively small cohort size,

144 RNAseq data from 130 TCGA PAAD samples was leveraged in order to increase sample size and enable

145 more robust clustering. Data were normalized using an empirical Bayesian method provided by

146 ComBat(21). Consensus clustering of z‐scores (of log10(x +1)) was performed, where x was RPKM for POG

147 samples and TPM for TCGA samples(5). Dendrograms were cut and branches labelled to assign subtypes

148 according to relative expression levels of subtyping gene sets.

149

150 Sequencing data availability

151 Genomic and transcriptomic datasets have been deposited at the European Genome‐phenome Archive

152 (EGA, http://www.ebi.ac.uk/ega/) under accession numbers EGAD00001004717 (patient 44),

153 EGAD00001004718 (patient 45) EGAD00001004716 (patient 46).

154

155 NRG1 fluorescent in situ hybridization (FISH) protocol

156 Fluorescent in situ hybridization was performed as previously described(2) using the Dako Histology FISH

157 Accessory Kit as per manufacturer’s protocol. Slides were scored manually using an oil immersion 63x

158 objective and z‐stack images were captured using Metasystems software (MetaSystems Group Inc.,

159 Belmont, MA, USA) (for details, see Supplemental Methods).

160

161 NRG1 fusion RT‐PCR

162 Total RNA was extracted from 6x5 micron scrolls of FFPE material with a High Pure FFPET RNA Isolation

163 kit (Roche, Mannheim, Germany). First strand cDNA was prepared from 200‐500ng RNA with standard

164 procedures for Superscript IV reverse transcriptase and random hexamer oligonucleotides (Thermo

165 Fisher Scientific, Waltham, Massachusetts, USA). Based on whole genome and transcriptome data,

166 oligonucleotide primers were designed to ATP1B1 exon 2 (FWD:5’‐CATCGGAACCATCCAAGTGA‐3’) and

167 exon 3 (FWD:5’‐TTTCGTCCTAATGATCCCAAGAG‐3’) and NRG1 exon 2 (RVS:5’‐

168 CATTCCCATTCTTGAACCACTTG‐3’) and exon 6 (RVS:5’‐CAGTGGTGGATGTAGATGTAGC‐3’). Each case was

169 assessed by standard Platinum Taq polymerase PCR protocol (Thermo Fisher Scientific, Waltham,

170 Massachusetts, USA) at 55oC annealing with ATP1B1ex2FWD‐NRG1ex2RVS and ATP1B1ex3FWD‐

171 NRG1ex6RVS primer combinations. Positive PCR bands were Sanger sequenced to confirm the fusion

172 breakpoint (NAPS, Michael Smith Laboratories, UBC, Vancouver, BC, Canada. For details, see

173 Supplemental methods).

174

175 TGCA fusion data review

176 Additional NRG1 gene fusions were identified by querying the Tumour Fusion Gene Data Portal

177 (http://www.tumorfusions.org)(22); which uses the PRADA algorithm(23) to detect fusion transcripts.

178

179 Results

180 Genomic and Transcriptomic Analysis

181 As part of the POG study, metastatic tumours from 47 patients with advanced PDAC underwent WGTA.

182 In agreement with previous genomic characterization studies of PDAC, clonal KRAS gain of function

183 mutations were identified in 91% (43/47) of tumour samples (Figure 1A, supplementary Table S1). KRAS

184 mutations were not identified in 4 samples by standard somatic single nucleotide variant analysis

185 (patients 44‐47) (supplementary Table S1). Subsequent targeted read alignment of known KRAS gain of

186 function hotspots revealed 1 of the 4 initially described KRAS wild‐type tumours (patient 47) had a KRAS

187 gain of function mutation of low variant allele frequency, indicating a potential subclonal KRAS mutation

188 (1 and 6 sequencing reads supporting a KRAS G12D in the genome and transcriptome, respectively).

189 Three of the 47 samples were devoid of KRAS mutations based on our standard and targeted SNV calling

190 protocols.

191

192 Structural variant (SV) analysis and annotation from genome and transcriptome sequencing data was

193 performed using MAVIS(15). Translocations affecting NRG1 were identified in 3 KRAS wild‐type tumour

194 samples (Figure 1A, supplementary Table S3). Notably, we did not detect oncogenic mutations or

195 structural variants in potentially actionable alternative drivers including BRAF, ERBB2, ROS1, ALK,

196 NTRK1‐3, BRCA1/2 or ATM. Furthermore, recurrent loss of function mutations and deletions affecting

197 tumour suppressors CDKN2A, TP53 and SMAD4 were absent in tumours positive for NRG1 fusions

198 (Figure 1A, supplementary Table S1 and S2). Additionally, expression‐based subtyping revealed that all

199 three KRAS wild‐type, NRG1 fusion positive samples were consistent with the classical subtype of PDAC

200 (Figure 1A, supplementary Table S4)(18). Somatic SNVs were predominantly C:G>T:A and T:A>C:G

201 transitions, similar to the majority of the cohort (Figure 1A), and were compared to a catalogue of

202 previously described mutation signatures (http://cancer.sanger.ac.uk/cosmic/signatures). Somatic SNVs

203 from all three NRG1 fusion positive tumours were most highly attributed to COSMIC mutation signature

204 1 and 5 which are ubiquitous across cancer types(24) (supplementary Table S5).

205

206 NRG1 Fusions in Advanced PDAC

207 Patient 44 is a 55‐year old man with a family history of GI cancers (gastric and colorectal cancer), who

208 presented with stage IV PDAC with liver metastases. No prior germline or somatic testing was

209 performed and he was enrolled on a first line metastatic phase II clinical trial of gemcitabine and nab‐

210 paclitaxel vs gemcitabine, nab‐paclitaxel and dual‐checkpoint inhibitor therapy. WGTA was performed

211 on a core needle biopsy from a metastatic lesion in the liver. Tumour content was estimated to be 40%

212 by pathology. Analysis of the somatic data revealed 42 somatic mutations (SNVs and indels)

213 (supplementary Table S6). The tumour was negative for KRAS mutations and none of the other

214 mutations were determined to be of clinical relevance. SV analysis revealed a fusion of ATP1B1 (exon 4)

215 with NRG1 (exon 6), that was supported by both the genome and transcriptome data (Figure 1B,

216 supplementary Table S3). Expression analysis of the NRG1 gene revealed the tumour sample from

217 patient 44 was a high expression outlier compared to a cohort of primary pancreatic adenocarcinoma

218 samples, suggesting increased expression of NRG1 as a result of the fusion transcript. Alignment of

219 RNAseq reads to the NRG1 gene demonstrated a significant increase in expression of exons 6 and 7

220 downstream of the predicted breakpoint, indicating the translocation stimulated expression of the

221 fusion transcript (Figure 2B). This patient remains on first line therapy.

222

223 Patient 45 is a 59‐year old man with a family history of prostate and colon cancer, who presented with

224 stage IV PDAC with multiple metastases to the liver. No prior germline or somatic testing was performed.

225 He was consented and enrolled in the POG clinical trial after having received 10 cycles of standard

226 treatment with oxaliplatin, irinotecan, and 5‐FU (FOLFIRINOX), with some initial partial response and

227 subsequent progression. He was subsequently treated with 3 cycles of gemcitabine with progression.

228 WGTA analysis performed on a biopsy sample (pathology‐estimated tumour content of 40%) from a

229 metastatic lesion from the liver showed 52 somatic mutations (SNVs and indels) (supplementary Table

230 S6), none of which were determined to be clinically relevant. SV analysis revealed an ATP1B1‐NRG1 gene

231 fusion that was supported by both the genome and transcriptome data (Figure 1B, supplementary Table

232 S3). The breakpoints identified in patient 45 resulted in fusion of exon 3 of ATP1B1 with exon 2 of NRG1,

233 and differed from those observed in patient 44. Similar to patient 44, NRG1 transcripts were highly

234 expressed in the tumour sample from patient 45, which was a significant outlier when compared to the

235 cohort of primary PDAC, as described above (Figure 2A). In agreement with the fusion breakpoints

236 detected in this sample, aligned RNAseq reads across the NRG1 gene revealed an increase in expression

237 of exons 2‐7 (Figure 2B). ATP1B1‐NRG1 fusions were subsequently validated by RT‐PCR followed by

238 Sanger sequencing.

239

240 Testing by break‐apart FISH did not confirm the presence of ATP1B1‐NRG1 fusions in patient 44 or

241 patient 45. Upon examination of the genome assembly data additional genomic breakpoints

242 downstream of the fusion junction were identified in each case. An analysis of the structural complexity

243 determined that the fusions likely arise from the insertion of the C‐terminal region of the NRG1 gene

244 from chromosome 8 into the ATP1B1 locus on chromosome 1 (supplementary Figure S1).

245

246 Patient 46 is a 54‐year old man who has a limited family history of cancer and presented with stage IV

247 PDAC with metastasis to the liver. The patient was consented and enrolled in the POG and PanGen

248 clinical trials and received gemcitabine with nab‐paclitaxel as part of a clinical trial. WGTA analysis of the

249 liver biopsy (pathology‐estimated tumour content of 85%) revealed 120 somatic mutations (SNVs and

250 indels) including homozygous frame shift mutations affecting ARID1A (p.Q1519fs) and B2M (p.S53fs) and

251 several variants of unknown significance affecting additional components of the antigen presenting class

252 I MHC complex, HLA‐A (p.V255M and p.D251H) and HLA‐B (p.T162K) (supplementary Table S6). The

253 tumour was negative for KRAS gain of function mutations. A complex structural rearrangement

254 resulting in the insertion of exons 6 and 7 of NRG1 in between exons 15 and 16 of APP was detected in

255 both the genome and transcriptome (Figure 1B, supplementary Table S3). Expression analysis indicated

256 the fusion transcript resulted in increased expression of exons 6 and 7 of the NRG1 gene and outlier

257 expression compared to the primary PDAC cohort (Figure 2). Similar to patients 44 and 45, insertion of

258 the NRG1 genomic fragment containing exons 6 and 7 into the APP gene were predicted to not be

259 detectable by break‐apart FISH.

260

261 All three NRG1 fusions were predicted to be in‐frame and the predicted fusion structure preserved the

262 EGF‐like domain, which has previously been shown to be required for NRG1 function(25, 26) (Figure 1B).

263 Moreover, alignment of the RNA sequencing reads revealed a striking increase in expression of EGF‐like

264 domain containing exons 6 and 7 in all three cases (Figure 2B). Both fusion partners, ATP1B1 and APP,

265 encode integral membrane proteins and donate trans‐membrane domains to the predicted NRG1 fusion

266 proteins (Figure 1B). Finally, a query of gene fusions detected in transcriptome data from TCGA

267 [http://www.tumorfusions.org/](22) identified a third ATP1B1‐NRG1 in‐frame gene fusion in a sample

268 from a PDAC (Table 1). The fusion junction results in the fusion of exon 3 from ATP1B1 and exon 2 of

269 NRG1, identical to the structure of the gene fusion identified in patient 45 (Figure 1B).

270

271 Clinical Response to Afatinib in NRG1 Fusion Positive PDAC

272 Following progression on gemcitabine, patient 45 obtained access to afatinib. At the time of starting

273 therapy, he had significant ascites and was ECOG performance status 2. His CA19‐9 was >120,000. A

274 PET/CT scan was performed at baseline prior to initiating therapy (Figure 3A, left panels). He initiated

275 treatment at a dose of 40 mg daily but was switched to 30 mg during cycle 1 due to diarrhea. After 4

276 weeks, a repeat PET/CT was performed demonstrating a significant response to therapy. It was noted

277 that the degree and extent of FDG avidity within supraclavicular, mediastinal, portocaval and liver had

278 all improved. Repeat CA19‐9 after 4 weeks dropped to 7,246 (Figure 3C). Clinically he appeared much

279 better with significant improvements in pain and performance status. He had minor skin rash and minor

280 diarrhoea related to the afatinib. He had a CT at 3 months post treatment initiation confirming ongoing

281 response. At 5.5 months since treatment initiation he had a repeat CT that demonstrated disease

282 progression.

283

284 Patient 46 did not tolerate gemcitabine and nab‐paclitaxel and came off treatment after 1 cycle. At the

285 time of starting afatinib he had significant progression of disease with PET/CT showing avidity in the

286 pancreatic head mass, as well as in multiple metastatic lymph nodes and liver metastasis (Figure 3B, left

287 panels). CA19‐9 levels were within the normal range. He initiated treatment with afatinib at a dose of 30

288 mg daily. After 4 weeks, a repeat PET/CT was performed also showing significant response to therapy

289 (Figure 3 B, right panels). There was noted to be resolution of multiple hepatic metastases, and the

290 degree of FDG avidity within the primary pancreatic mass had also decreased. He tolerated the

291 treatment well with good energy and minimal toxicity other than minor facial rash and paronychia.

292 Response was confirmed by CT imaging performed at 8 weeks post treatment initiation, and a CT scan

293 done 5 months post treatment initiation showed ongoing response. He continues on treatment at the

294 time of manuscript preparation.

295 Discussion

296 In this report, we describe ATP1B1‐NRG1 gene fusions in 2 out of the 47 PDAC samples sequenced as

297 part of the ongoing POG trial at our institute. Within this cohort, 3 PDAC samples were KRAS wild‐type

298 and harboured NRG1 fusions. A fourth was identified in a review of publically available transcriptome

299 and 3 additional cases were recently described in independent study data(6). Together these results

300 indicate that NRG1 fusions occur in a small proportion of PDAC patients, potentially defining a new

301 subtype of KRAS wild‐type PDAC.

302

303 NRG1 activates the ERBB3 (HER3) receptor, which then forms a heterodimer with other HER‐family

304 receptors, regulating downstream signaling pathways. NRG1 gene fusions are prevalent in lung cancers

305 and are thought to drive ectopic signaling through continuous stimulation of HER‐family proteins leading

306 to growth and proliferation(27, 28). It is then not surprising that NRG1 gene fusions could drive PDAC

307 given that the majority of PDAC is driven by KRAS activation, providing continuous activation of signaling

308 pathways including MEK, ERK and PI3K(29). This hypothesis is borne out by the emerging mutual

309 exclusivity of NRG1 gene fusions and KRAS activating mutations, which suggests an overlapping

310 functions. The absence of potential alternative drivers of PDAC in NRG1 fusion positive tumours

311 provides further support for NRG1 fusions as a driver in PDAC.

312

313 Two patients were treated with afatinib as a result of the NRG1 fusions detected as part of this study.

314 Both patients demonstrated a significant clinical response when treated with afatinib. This observation

315 is interesting given that a recent publication described an observable lack of response in some lung

316 cancers positive for NRG1 fusions(4). This perhaps suggests that the fusion partner or etiology has an

317 influence on how well a given tumour will respond. Alternatively, the genomic context of the gene

318 fusion may be the relevant factor indicating the importance of comprehensive genomic profiling to

319 determine if the gene fusions detected are likely to be to sole driver of the disease. Further work is

320 required to explore these observations.

321

322 Patient 45 showed disease progression after 5.5 months on treatment, a response to afatinib broadly

323 equivalent to that seen in an unselected population of patients with advanced NSCLC(30). This

324 observation demonstrates the likelihood of resistance mechanisms arising in pre‐treated NRG1 fusion‐

325 positive PDAC. Up‐regulation of NRG1 is a well‐documented mechanism of parallel pathway activation

326 and resistance in HER2‐positive breast cancer models(31, 32) and ALK‐positive lung cancer(33, 34).

327 Furthermore, it has been reported that PDAC cells treated with an ERK‐specific inhibitor upregulated the

328 parallel PI3K‐AKT pathway through activating HER‐ family proteins(35). It is therefore perhaps likely that

329 resistance mechanisms in HER‐family driven PDAC will emerge from the acquisition of genomic

330 alterations that activate parallel signaling pathways.

331

332 With respect to possible therapeutic strategies a recent study has demonstrated that concurrent

333 inhibition or HER‐family kinases synergizes with an ERK‐specific inhibitor in suppressing PDAC cell

334 growth in vitro and in vivo(35). This provides a rationale for testing combinations of targeted therapies

335 in preclinical models, as has been demonstrated for cancers treated with BRAF inhibitors(36, 37). Also,

336 given that NRG1 is the primary ligand of ERBB3(38) there is significant interest in targeting ERBB3

337 directly. A recent study using breast cancer cell lines harbouring an NRG1 gene fusion and a model of

338 NRG1 fusion‐positive ovarian cancer revealed potential complexities in responses to broad versus

339 targeted HER‐familiy kinase inhibitors in NRG1‐fusion positive disease(4). However, establishing an

340 NRG1 positive model of PDAC will be instrumental in investigating appropriate treatments strategies for

341 this etiology. To date, no ERBB3‐targeted therapy has been approved for cancer treatment though

342 monoclonal antibodies are under clinical evaluation(39, 40) and the number of phase I and phase II

343 clinical trials continues to grow. However, given the rarity and structural complexity of NRG‐1 fusions,

344 developing targeted strategies to reliably detect these genomic events will be necessary to facilitate the

345 appropriate interpretation of clinical trial outcomes.

346

347 Finally, in addition to a previously described SDC4‐NRG1 gene fusion in a case of lung cancer two

348 additional in‐frame NRG1 gene fusions were identified in transcriptome data from TCGA; CDK1‐NRG1 in

349 a bladder urothelial carcinoma and PDE7A‐NRG1 in a head and neck squamous cell carcinoma (Table 1).

350 Though the evidence is currently limited, this finding suggests that broad screening for NRG1 gene

351 fusions across different cancer types may reveal additional potentially clinically actionable oncogenic

352 events that can be tested for their therapeutic relevance in a clinical or pre‐clinical setting.

353 Acknowledgements

354 We gratefully acknowledge the participation of our patients and families, the POG and PanGen team,

355 and the generous support of the BC Cancer Foundation and Genome British Columbia. The results

356 published here are in part based upon data generated by The Cancer Genome Atlas managed by the NCI

357 and NHGRI (http://cancergenome.nih.gov). MRJ had full access to all the data in the study and takes

358 responsibility for the integrity of the data and the accuracy of the data analysis. MRJ, LMW (BC Cancer,

359 Canada’s Michael Smith Genome Sciences Centre) and JTT (Pancreas Centre BC, Vancouver, Canada)

360 conducted and are responsible for the data analysis. Thank you to Dr. Daniel F. Worsley and Dr. R. Petter

361 Tonseth for their help in obtaining the radiological images.

362

363 Clinical trial information

364 Personalized OncoGenomics (POG) Program of British Columbia: Utilization of Genomic Analysis to

365 Better Understand Tumour Heterogeneity and Evolution (NCT02155621); Prospectively Defining

366 Metastatic Pancreatic Ductal Adenocarcinoma Subtypes by Comprehensive Genomic Analysis (PanGen;

367 NCT02869802).

368

369 References:

370 1. Gay ND, et al. (2017) Durable Response to Afatinib in Lung Adenocarcinoma Harboring 371 NRG1 Gene Fusions. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer 12(8):e107– 372 e110.

373 2. Jones MR, et al. (2017) Successful targeting of the NRG1 pathway indicates novel treatment 374 strategy for metastatic cancer. Ann Oncol 28(12):3092–3097.

375 3. Cheema PK, Doherty M, Tsao M-S (2017) A Case of Invasive Mucinous Pulmonary 376 Adenocarcinoma with a CD74-NRG1 Fusion Protein Targeted with Afatinib. J Thorac 377 Oncol Off Publ Int Assoc Study Lung Cancer 12(12):e200–e202.

378 4. Drilon A, et al. (2018) Response to ERBB3-Directed Targeted Therapy in NRG1-Rearranged 379 Cancers. Cancer Discov 8(6):686–695.

380 5. Cancer Genome Atlas Research Network. Electronic address: 381 [email protected], Cancer Genome Atlas Research Network (2017) 382 Integrated Genomic Characterization of Pancreatic Ductal Adenocarcinoma. Cancer Cell 383 32(2):185–203.e13.

384 6. Heining C, et al. (2018) NRG1 Fusions in KRAS Wild-Type Pancreatic Cancer. Cancer 385 Discov 8(9):1087–1095.

386 7. Saunders CT, et al. (2012) Strelka: accurate somatic small-variant calling from sequenced 387 tumor-normal sample pairs. Bioinforma Oxf Engl 28(14):1811–1817.

388 8. Ding J, et al. (2012) Feature-based classifiers for somatic mutation detection in tumour- 389 normal paired sequencing data. Bioinforma Oxf Engl 28(2):167–175.

390 9. Ha G, et al. (2012) Integrative analysis of genome-wide loss of heterozygosity and 391 monoallelic expression at nucleotide resolution reveals disrupted pathways in triple- 392 negative breast cancer. Genome Res 22(10):1995–2007.

393 10. Robertson G, et al. (2010) De novo assembly and analysis of RNA-seq data. Nat Methods 394 7(11):909–912.

395 11. McPherson A, et al. (2011) deFuse: an algorithm for gene fusion discovery in tumor RNA- 396 Seq data. PLoS Comput Biol 7(5):e1001138.

397 12. Rausch T, et al. (2012) DELLY: structural variant discovery by integrated paired-end and 398 split-read analysis. Bioinforma Oxf Engl 28(18):i333–i339.

399 13. Iyer MK, Chinnaiyan AM, Maher CA (2011) ChimeraScan: a tool for identifying chimeric 400 transcription in sequencing data. Bioinforma Oxf Engl 27(20):2903–2904.

401 14. Chen X, et al. (2016) Manta: rapid detection of structural variants and indels for germline 402 and cancer sequencing applications. Bioinforma Oxf Engl 32(8):1220–1222.

403 15. Reisle C, et al. (2018) MAVIS: Merging, Annotation, Validation, and Illustration of 404 Structural variants. Bioinforma Oxf Engl. doi:10.1093/bioinformatics/bty621.

405 16. Cancer Genome Atlas Research Network, et al. (2013) The Cancer Genome Atlas Pan- 406 Cancer analysis project. Nat Genet 45(10):1113–1120.

407 17. Alexandrov LB, Nik-Zainal S, Wedge DC, Campbell PJ, Stratton MR (2013) Deciphering 408 signatures of mutational processes operative in human cancer. Cell Rep 3(1):246–259.

409 18. Moffitt RA, et al. (2015) Virtual microdissection identifies distinct tumor- and stroma- 410 specific subtypes of pancreatic ductal adenocarcinoma. Nat Genet 47(10):1168–1178.

411 19. Bailey P, et al. (2016) Genomic analyses identify molecular subtypes of pancreatic cancer. 412 Nature 531(7592):47–52.

413 20. Collisson EA, et al. (2011) Subtypes of pancreatic ductal adenocarcinoma and their 414 differing responses to therapy. Nat Med 17(4):500–503.

415 21. Johnson WE, Li C, Rabinovic A (2007) Adjusting batch effects in microarray expression 416 data using empirical Bayes methods. Biostat Oxf Engl 8(1):118–127.

417 22. Hu X, et al. (2018) TumorFusions: an integrative resource for cancer-associated transcript 418 fusions. Nucleic Acids Res 46(D1):D1144–D1149.

419 23. Torres-García W, et al. (2014) PRADA: pipeline for RNA sequencing data analysis. 420 Bioinforma Oxf Engl 30(15):2224–2226.

421 24. Alexandrov LB, et al. (2013) Signatures of mutational processes in human cancer. Nature 422 500(7463):415–421.

423 25. Jung Y, et al. (2015) VAMP2-NRG1 Fusion Gene is a Novel Oncogenic Driver of Non- 424 Small-Cell Lung Adenocarcinoma. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer 425 10(7):1107–1111.

426 26. Guo R, et al. (2016) Discovery of ERBB3 inhibitors for non-small cell lung cancer 427 (NSCLC) via virtual screening. J Mol Model 22(6):135.

428 27. Saito M, et al. (2016) Gene aberrations for precision medicine against lung adenocarcinoma. 429 Cancer Sci 107(6):713–720.

430 28. Jeong H, et al. (2014) Neuregulin-1 induces cancer stem cell characteristics in breast cancer 431 cell lines. Oncol Rep 32(3):1218–1224.

432 29. Polireddy K, Chen Q (2016) Cancer of the Pancreas: Molecular Pathways and Current 433 Advancement in Treatment. J Cancer 7(11):1497–1514.

434 30. Ezeife DA, et al. (2018) Afatinib in advanced pretreated non-small-cell lung cancer- a 435 Canadian experience. Curr Oncol Tor Ont 25(5):e385–e390.

436 31. Xia W, et al. (2013) An heregulin-EGFR-HER3 autocrine signaling axis can mediate 437 acquired lapatinib resistance in HER2+ breast cancer models. Breast Cancer Res BCR 438 15(5):R85.

439 32. Prentice LM, et al. (2005) NRG1 gene rearrangements in clinical breast cancer: 440 identification of an adjacent novel amplicon associated with poor prognosis. Oncogene 441 24(49):7281–7289.

442 33. Dong X, Fernandez-Salas E, Li E, Wang S (2016) Elucidation of Resistance Mechanisms to 443 Second-Generation ALK Inhibitors Alectinib and Ceritinib in Non-Small Cell Lung Cancer 444 Cells. Neoplasia N Y N 18(3):162–171.

445 34. Kimura M, et al. (2015) Analysis of ERBB ligand-induced resistance mechanism to 446 crizotinib by primary culture of lung adenocarcinoma with EML4-ALK fusion gene. J 447 Thorac Oncol Off Publ Int Assoc Study Lung Cancer 10(3):527–530.

448 35. Jiang H, et al. (2018) Concurrent HER or PI3K Inhibition Potentiates the Antitumor Effect 449 of the ERK Inhibitor Ulixertinib in Preclinical Pancreatic Cancer Models. Mol Cancer Ther 450 17(10):2144–2155.

451 36. Xue Y, et al. (2017) An approach to suppress the evolution of resistance in BRAFV600E- 452 mutant cancer. Nat Med 23(8):929–937.

453 37. Firestone AJ, Settleman J (2017) A three-drug combination to treat BRAF-mutant cancers. 454 Nat Med 23(8):913–914.

455 38. Mujoo K, Choi B-K, Huang Z, Zhang N, An Z (2014) Regulation of ERBB3/HER3 456 signaling in cancer. Oncotarget 5(21):10222–10236.

457 39. Shimizu T, et al. (2017) Phase 1 study of new formulation of patritumab (U3-1287) Process 458 2, a fully human anti-HER3 monoclonal antibody in combination with erlotinib in Japanese 459 patients with advanced non-small cell lung cancer. Cancer Chemother Pharmacol 460 79(3):489–495.

461 40. Karachaliou N, Lazzari C, Verlicchi A, Sosa AE, Rosell R (2017) HER3 as a Therapeutic 462 Target in Cancer. BioDrugs Clin Immunother Biopharm Gene Ther 31(1):63–73.

463

464 Figure 1. NRG1 fusions detected in KRAS wild‐type PDAC. A. Molecular alteration status of genes

465 recurrently altered in PDAC and gene expression‐based molecular subtyping in the advanced PDAC

466 cohort (n=47). NRG1 fusions were detected in three KRAS wild‐type patients. B. Fusion visualization

467 diagrams for NRG1 gene fusions in KRAS wild‐type patient tumours. Exons corresponding to NRG1

468 (ENST00000523534), ATP1B1 (ENST00000367816) and APP (ENST00000346798) incorporated in the

469 fusion transcripts are indicated. Reading frame is indicated by the black bar below the fusion transcript

470 model and EGF‐like and transmembrane domains derived from NRG1 and the fusion partners

471 respectively are highlighted by the dark blue and beige boxes.

472 Figure 2. Expression of NRG1 fusion transcripts. A. Histogram of NRG1 expression (log10(RPKM)) across

473 the primary TCGA pancreatic cohort with expression of NRG1 fusion positive PDAC patients indicated by

474 the dashed lines. B. Density of RNAseq read alignment across the NRG1 transcript (ENST00000523534)

475 for NRG1 fusion positive PDAC patient tumours.

476 Figure 3. Clinical response of NRG1 fusion positive PDAC with afatinib. A. PET/CT scans of patient 45

477 prior to treatment (left) and after 4 weeks on afatinib (right). B. PET/CT scans of patient 46 prior to

478 treatment (left) and after 4 weeks on afatinib (right). C. CA19‐9 levels pre and post‐afatinib for patient

479 45.

Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2019 American Association for Cancer Research. Downloaded from Table 1: In‐frame ATP1B1‐NRG1 gene fusions detected in transcriptome data from TCGA and POG and described in published literature Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. Author ManuscriptPublishedOnlineFirstonMay8,2019;DOI:10.1158/1078-0432.CCR-19-0191 Cancer Type Sample Fusion‐Pair Frame 5' Gene Junction 3' Gene Junction 5' exon 3' exon

clincancerres.aacrjournals.org PDAC POG‐Patient 44 ATP1B1‐NRG1 In‐frame Chr1:169094277/1 Chr8:32585467/1 4 6

PDAC POG‐Patient 45 ATP1B1‐NRG1 In‐frame Chr1:169080736/1 Chr8:32453346/1 3 2

PDAC POG‐Patient 46 APP‐NRG1 In‐frame Chr21:27277336/1 Chr8:32585467/1 15,7 6,16 Chr8:32600251/1 Chr21:27269985/1

PDAC TCGA.3A.A9I5.01A ATP1B1‐NRG1 In‐frame Chr1:169080736/1 Chr8:32453346/1 3 2

PDAC NCT/DKTK MASTER(6) ATP1B1‐NRG1 In‐frame Chr1:169080736/1 nd 3 2

BLCA TCGA.FD.A5BR.01A CDK1‐NRG1 In‐frame Chr10:62539960/1 Chr8:32585467/1 2 6

HNSC TCGA.CQ.5334.01A PDE7A‐NRG1 In‐frame Chr8:66691955/‐1 Chr8:32585467/1 3 6

LUAD TCGA.86.7954.01A SDC4‐NRG1 In‐frame Chr20:43959006/‐1 Chr8:32585467/1 4 6

LUAD POG‐patient 4(2) SDC4‐NRG1 In‐frame Chr20:43959006/‐1 Chr8:32585467/1 4 6

25 Downloaded from PDAC = Pancreatic Ductal Adenocarcinoma, CHOL = Cholangiocarcinoma, BLCA = bladder urothelial carcinoma, HNSC = head and neck squamous Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited.

cell carcinoma, LUAD = lung adenocarcinoma, nd = no data Author ManuscriptPublishedOnlineFirstonMay8,2019;DOI:10.1158/1078-0432.CCR-19-0191

26 Author Manuscript Published OnlineFirst on May 8, 2019; DOI: 10.1158/1078-0432.CCR-19-0191 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 1

A 600

400

200 Coding Mutation Count

0 1.00 substitution 0.75 C:G>A:T C:G>G:C 0.50 C:G> T:A T:A>A:T 0.25 T:A>C:G T:A>G:C 0.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 variant KRAS deletion NRG1 amplification truncating TP53 missense CDKN2A translocation SMAD4 GNAS ERBB2 ALK MYC FGFR1 BRCA2 ATM Subtype MSH6 basal-like classical MSH3 na RNF43 classical exo-like quasimes Moffitt Collison adex immuno Bailey progen squamous

B Patient 44

ATP1B1-NRG1

Patient 45

ATP1B1-NRG1

Patient 46

APP-NRG1

Freq uency 0 10 20 30 40 50 − 3 Downloaded from Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. − 2 Author ManuscriptPublishedOnlineFirstonMay8,2019;DOI:10.1158/1078-0432.CCR-19-0191 log clincancerres.aacrjournals.org − 1 10 ( RP K M) pt.46 0 pt.45 1 on September 30, 2021. © 2019American Association for Cancer Research. pt.44 2 Author Manuscript Published OnlineFirst on May 8, 2019; DOI: 10.1158/1078-0432.CCR-19-0191 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 3 A

pre-treatment + 4 weeks afatinib

C baseline post-afatinib

100000

50000 CA19-9 level

0 0 1 2 Downloadedtime (months) from clincancerres.aacrjournals.org on September 30, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on May 8, 2019; DOI: 10.1158/1078-0432.CCR-19-0191 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

NRG1 gene fusions are recurrent, clinically actionable gene rearrangements in KRAS wild-type pancreatic ductal adenocarcinoma

Martin R Jones, Laura M Williamson, James T. Topham, et al.

Clin Cancer Res Published OnlineFirst May 8, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-19-0191

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/05/08/1078-0432.CCR-19-0191.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2019/05/08/1078-0432.CCR-19-0191. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.