Genomic mechanisms of resistance to tyrosine kinase inhibitors in HER2 amplified breast cancer

doi:10.21203/rs.3.rs-4270758/v1

Download PDF

Article

Genomic mechanisms of resistance to tyrosine kinase inhibitors in HER2 amplified breast cancer

https://doi.org/10.21203/rs.3.rs-4270758/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Though there has been substantial progress in the development of anti-HER2 therapies to treat HER2-positive metastatic breast cancer (MBC) within the past two decades, most patients still experience disease progression and cancer-related death. HER2-directed tyrosine kinase inhibitors (TKIs) can be highly effective therapies for patients with HER2-positive MBC, however, an understanding of resistance mechanisms is needed to better inform treatment approaches. We performed whole exome sequencing on 111 patients with 73 tumor biopsies and 120 cell-free DNA (cfDNA) samples to assess mechanisms of resistance. In 11/26 patients with acquired resistance, we identified alterations in previously characterized genes, such as PIK3CA and ERBB2 that could explain treatment resistance. Mutations in growing subclones identified potential novel mechanisms of resistance in 5/26 patients and included alterations in ESR1, FGFR2, and FGFR4. Additional studies are needed to assess the functional role and clinical utility of these alterations in driving resistance.

Biological sciences/Cancer/Cancer genomics

Biological sciences/Cancer/Breast cancer

Human epidermal growth factor receptor 2 expressing (HER2-positive or ERBB2 amplified) tumors account for 15–20% of breast cancers. HER2-positive breast cancers are associated with high-grade histology, higher rates of recurrence, and higher occurrence of brain metastases. In the first-line setting, combining two HER2-targeted antibodies, trastuzumab and pertuzumab, with chemotherapy is the current standard of care regimen for patients with HER2-positive, metastatic breast cancer (MBC).^1,2 In later-line settings, HER2-targeted tyrosine kinase inhibitors (TKIs) can be effective and are a widely utilized approach.³ However, most patients eventually develop treatment resistance.

TKIs, a class of small molecules targeting the intracellular kinase domain of receptor tyrosine kinases such as HER2, have been shown to improve patient outcomes in numerous tumor subtypes.⁴ The most common HER2-directed TKIs compete with ATP binding and block substrate phosphorylation, thus preventing the activation of downstream signaling pathways. Many TKIs - including lapatinib, neratinib and tucatinib - have been approved for use in patients with HER2-positive MBC and have particular benefit in patients with brain metastases.³ While many patients experience clinically meaningful responses, most tumors will eventually develop resistance to TKIs leading to disease progression. Preclinical studies in mice xenografts and in vitro human breast cancer cell lines have shown that activating PIK3CA mutations confer resistance to lapatinib.⁵ Prior work has also demonstrated that L755S, a HER2 kinase domain mutation, leads to lapatinib resistance and the activation of the downstream MAPK and PI3K/AKT/mTOR pathways.⁶ While many resistance mechanisms have been demonstrated in the preclinical setting, there is limited knowledge about these mechanisms in patients.

Patients receiving HER2-directed TKIs, sampled by cell-free DNA (cfDNA) and tumor biopsy before, during and after therapy, present a unique opportunity to examine genomic alterations that may explain therapeutic resistance. In this investigation of the in vivo genomic landscape of mechanisms of resistance within a cohort of patients with HER2-positive MBC, we report both acquired and intrinsic resistance mechanisms to three HER2-directed tyrosine kinase inhibitors (lapatinib, neratinib, and tucatinib).

Of 111 participants analyzed, 26 – the “paired cohort” – had samples taken prior to and after ≥ 6 weeks of TKI treatment; 30 – the “post-TKI cohort” – had samples collected only after ≥ 6 weeks TKI treatment; and 55 – the “pre-TKI cohort” – had samples collected only prior to ≥ 6 weeks TKI treatment (Fig. 1A). All participants across the three cohorts had HER2-positive breast cancer per American Society of Clinical Oncology - College of American Pathology guidelines at time of metastatic diagnosis.⁷ All paired cohort participants were HER2-positive at time of primary diagnosis except for one participant who did not undergo HER2 testing due to lack of availability of HER2 testing at the time. Detailed primary and metastatic receptor status for the paired cohort is included in Supplementary Tables 1 and 2, respectively. Median age at metastatic diagnosis for the paired cohort was 48 years. Most participants initially presented with stage I-III disease (66%) with the remaining presenting as de novo metastatic (34%). Participants were followed for a median of 4.9 years from metastatic diagnosis and the majority were female (96%) with hormone receptor-positive (HR-positive) disease (62%). Most participants had visceral disease (69%), and 65% of participants had central nervous system (CNS) involvement. Most participants in the paired cohort received lapatinib (92%) and a median number of 6 (range 1–10) lines of metastatic therapy. Median overall survival in the paired cohort was 4.51 years from metastatic diagnosis. A median number of 3 samples (2–7) per participant were collected, with a median tumor fraction of 13.2% (2.8%-53.2%). The post-TKI and pre-TKI cohorts were comparable to the paired cohort for median age at diagnosis, overall survival, follow-up time, gender, hormone receptor (HR) status, and race (Supplementary Table 3).

The genomic landscape of post-TKI HER2-positive advanced breast cancer

To investigate mechanisms underlying resistance to HER2-targeted TKIs, we performed whole exome sequencing (WES) on 192 tumor and cfDNA samples from 111 participants with acquired or intrinsic resistance to anti-HER2 TKIs. Alterations in previously characterized genes that could explain treatment resistance were labeled as known mechanisms, including kinase domain mutations in ERBB2 (L755S, D769Y, V777L) and activating PIK3CA mutations.^5,6,8 Potential mechanisms were identified in genes with mutations known to cause resistance in other cancer types or therapies (Fig. 1A).

In the paired cohort, we found that eleven (42%) participants had at least one known alteration leading to resistance and 38% had no identified mechanism of resistance (Fig. 1B). We then assessed the frequency of alterations in canonical oncogenic pathways, as well as a gene set containing ESR1 and its transcriptional regulators (Supplementary Table 4), to determine whether any pathways were significantly recurrently mutated in the cohort.^9,10 The full list of genes altered in each pathway can be found in Supplementary Table 5 with variant classification for each mutation. RTK-RAS and TP53 were the most frequently mutated pathways with 16 (62%) participants having an alteration, followed by the WNT pathway (11 participants, 42%) and the HIPPO, NOTCH, and ESR1 and regulators pathways (each 10 participants, 38%). The TP53 and ESR1 and regulators pathways were both significantly mutated (q < 0.05) above the expected background mutation rate. We analyzed known mutational signatures and found the most frequently represented signatures were clock-like, APOBEC and capecitabine (Fig. 1B).¹¹ We identified no other statistically significant differences at a pathway level between cohorts (Fig. 1C).

In the pre-TKI cohort, RTK-RAS was the most frequently mutated pathway with 31 (56%) participants having an alteration, followed by the TP53 pathway (30 participants, 55%) and PI3K pathway (25 participants, 45%) (Supplementary Fig. 1A). In the post-TKI cohort, TP53 was the most frequently mutated pathway with 23 (77%) participants having an alteration, followed by the PI3K pathway (19 participants, 63%) and RTK-RAS pathway (18 participants, 60%) (Supplementary Fig. 1B). The TP53 pathway was significantly mutated (q < 0.05) in both the pre- and post-TKI cohorts, and the PI3K pathway was significantly mutated (q < 0.05) in the post-TKI cohort only.

Mutation frequencies in the most frequently altered genes were similar between the paired cohort and the pre- and post-TKI cohorts. TP53, the most frequently altered gene in all three cohorts, was altered in 58% of paired participants, 67% of post-TKI participants, and 53% of pre-TKI participants. PIK3CA was altered in 31% of paired participants, 50% of post-TKI participants, and 35% of pre-TKI participants. ERBB2 was altered in 19% of paired participants, 17% of post-TKI participants, and 13% of pre-TKI participants. ESR1 mutations were enriched in the paired cohort (19%) compared to both the post-TKI cohort (10%) and pre-TKI cohort (5%) (Supplementary Table 6).

Next, we evaluated copy number alterations. The most frequently altered genes were ERBB2—as expected—followed by TP53, PTK2 and MYC, respectively. Six of eight participants in the paired acquired resistance cohort had focal high amplification of ERBB2, compared to only eight of the eighteen participants in the paired intrinsic cohort (p = 0.22, Fisher’s Exact test). TP53 biallelic inactivation—i.e., loss-of-function point mutation of one allele and deletion of the other allele—was observed across the paired cohort and occurred in 14 (54%) participants. Seven participants showed MYC amplifications, with one participant demonstrating focal high amplification. Five participants with intrinsic resistance did not have ERBB2 amplification (Fig. 2A), and lower pre-TKI ERBB2 copy number was associated with intrinsic resistance (p = 0.0074, Wilcoxon rank-sum test) (Fig. 2C). All five participants had clinical HER2-positive breast cancer confirmed on metastatic biopsy, a median of 842 days prior to assessed sample collection. The ERBB2 copy number of a sample was not associated with its genomic purity or tumor fraction (Spearman’s rho=-0.082, p = 0.46), suggesting that absence of ERBB2 amplification was not a result of low purity or tumor fraction. Within the paired cohort, ERBB2 copy number did not significantly change at a within-participant level during TKI treatment, and within-participant change did not differ significantly by HR status (Fig. 2B). When stratifying the paired cohort by HR status, participants with HR-positive MBC had lower ERBB2 copy number than HR-negative participants in both baseline and end-of-TKI treatment samples (p = 0.031 and p = 0.041, Wilcoxon rank-sum test), as has previously been shown (Supplementary Fig. 2).¹²

ESR1 alterations and ERBB2 copy number in HR-positive/HER2-positive MBC

Five out of the sixteen paired cohort participants with HR-positive disease had at least one activating ESR1 hotspot mutation (D538G or Y537S). Two of these five participants had never received an aromatase inhibitor (AI) or prior hormone therapy in either the early or metastatic setting. An activating ESR1 mutation presented in a growing subclone in four out of five participants. Three participants with an activating ESR1 mutation did not show ERBB2 copy number amplification per our analysis (Fig. 2A). Participants with an ESR1 mutation had significantly lower ERBB2 copy number relative to ESR1 wildtype cases in pre- and post-TKI paired samples (p = 0.024 and p = 0.0096, Wilcoxon rank-sum test) (Supplementary Fig. 3). When stratifying by HR status in an integrated analysis of the paired cohort and the pre- and post-TKI cohorts, participants with an activating ESR1 mutation had significantly lower ERBB2 copy number than both HR-negative and HR-positive ESR1 wildtype participants in both pre- and post-TKI samples (Fig. 2D).

Evolution of Clonal Dynamics in TKI resistance

To assess mechanisms of resistance we performed deeper evolutionary analysis on six participants with at least three serial samples collected (Fig. 3A). Known drivers of resistance were observed in multiple clones and included ERBB2 and PIK3CA mutations (Figs. 3B, 3C, 3D). Potential other mechanisms of resistance were identified as mutations in ESR1, FGFR2, and FGFR4 (Figs. 3B, 3E, 3F).

In participant 1, we observed two competing ESR1 mutations in separate clones with the clone harboring a Y537S mutation, a potential driver of resistance, initially predominant during lapatinib treatment. Upon treatment switch to capecitabine with trastuzumab and eventually trastuzumab with letrozole, we observed the clone containing the ESR1 D538G mutation outcompeting the Y537S-containing clone (Fig. 3B). One participant with a PIK3CA mutation also had several missense mutations, including a truncal ESR1 regulator pathway mutation present in PGR, a missense mutation in FOXA1, and a subclone with a NOTCH2 mutation growing during neratinib treatment (Fig. 3C). In another participant, growing subclones were detected with missense mutations in RTK-RAS, NOTCH, and WNT pathway genes such as SPRED2, EP300, and TLE2, respectively (Fig. 3G). However, for this participant, we did not observe any known or potential mechanisms of resistance, even though all participants were HER2 amplified (biopsy confirmed within 30 days of initial metastatic diagnosis) (Fig. 3G).

WNT pathway enrichment in HER2-positive brain metastases

Thirteen (50%) participants in the paired cohort developed brain metastases. These participants more often had HR-negative breast cancer (54% of participants who developed brain metastases vs. 23% of participants who did not; p = 0.23, Fisher’s Exact test). There were no significant differences in pathway mutation frequencies between participants who developed brain metastases and those who did not. However, mutations occurring specifically in the tumor suppressor genes of the WNT pathway were significantly enriched in participants with brain metastases (p = 0.0077, Boschloo’s test), with altered genes including APC, CHD8, LZTR1, TLE1, TLE2, and ZNRF3. Three of the five participants presented with subclonal WNT alterations in a cfDNA, breast or bone sample, with a median time of 903 days before the identification of any brain metastases (Fig. 4A, 4B). Evolutionary analysis for Patient 8, with HR-negative breast cancer brain metastasis, illustrates WNT pathway involvement in the growing subclone containing a missense mutation in the tumor suppressor gene TLE2 (Fig. 3G). In a combined analysis of the mutually exclusive pre- and post-TKI cohorts, mutations in WNT tumor suppressor genes remained significantly enriched in participants who ever developed brain metastases (p = 0.037, Boschloo’s test) (Supplementary Fig. 4). Of the eight participants with WNT tumor suppressor gene mutations in the combined pre- and post-TKI cohorts, two had the TSG alteration present before presentation of any brain metastases (median time of 1528 days before).

Though HER2-positive breast cancers predominantly remain dependent on the HER2 pathway, resistance develops in most metastatic tumors. Via detailed treatment data and sampling across the metastatic clinical course, we evaluated intrinsic and acquired resistance mechanisms in this cohort.

Beyond the established mechanisms - PI3K, ERBB2 and MAPK-pathway alterations - our study identified potential mechanisms of resistance driven by ESR1, FGFR2 and FGFR4 mutations. The other somatic alterations identified have biological potential to act as resistance mechanisms due to mutation type, location, and downstream signaling effects within canonical signaling pathways. However, unlike resistance to anti-EGFR TKIs in lung cancer, where approximately 50% of tumors develop acquired resistance via alteration of a single gene - EGFR - we observed a minority of participants (18%) with ERBB2 mutations in all post-TKI samples.¹³ These data highlight the diversity of resistance mechanisms and underscore that overcoming resistance will likely require multiple approaches. Given that we did not identify a common acquired resistance mechanism across the cohort and across different TKIs, sequential anti-HER2 TKI therapy may also be a viable treatment strategy. Further exploration in additional, larger cohorts with functional testing of mutations identified is warranted.

While less is known about WNT signaling in HER2-positive breast cancer, we observed WNT pathway alterations in well-established tumor suppressor genes in participants who had developed brain metastasis and validated this finding in an independent cohort of participants. WNT pathway alterations have previously been shown to mediate brain metastasis in lung cancer and activation of WNT signaling in glioblastoma is a known resistance mechanism to the established first-line treatment, temozolomide.^14,15 WNT pathway-mediated resistance is thought to act through over-expression of canonical ligands such as Wnt1 or through loss-of-function of tumor suppressors such as APC and has been shown to play a role in endocrine resistance of estrogen receptor-positive breast cancer.^16,17 Additional functional studies will be essential to further validate and characterize our intriguing finding of enrichment of WNT tumor suppressor gene mutations in patients with HER2 + breast cancer brain metastases.

All participants included in the study had biopsy proven clinical HER2 amplification and received anti-HER2 therapy. Others have shown that HR-positive, HER2-positive breast cancers have, on average, lower HER2 copy number than HR-negative, HER2-positive tumors.¹² In our study, participants with HR-positive, HER2-positive cancers and an ESR1 mutation were observed to have significantly lower HER2 copy number. In three of these participants, HER2 copy number was non-amplified per our genomic analysis. ESR1 ligand binding domain mutations occur in up to 40% of patients and are a well-established mechanism of resistance to aromatase inhibitors in HR-positive breast cancers.¹⁸ESR1 mutations are rare in treatment-naive tumors and are associated with estrogen pathway dependence. Pre-clinical evidence suggests that crosstalk between the HER2 and ER pathways mediates HER2 resistance, and that dual blockade may be important in effective therapy.^19–21 Our data provide further evidence of dependency on this pathway. Targeting these two pathways simultaneously was explored in the CALGB 40302 trial with clinical benefit seen with the combination of fulvestrant and lapatinib for patients with HR-positive/HER2-positive tumors.²² However, current guidelines recommend treating HER2-positive MBC with HER2-directed therapy regardless of ER status. Our data raise the question of whether HER2-positive/ER-positive breast cancer may be not one, but at least two entities, with varying dependence on these two pathways. With the development of additional anti-estrogen therapies, such as novel selective estrogen receptor degraders (SERDs), complete estrogen receptor antagonists (CERANs), selective estrogen receptor covalent antagonists (SERCAs), and proteolysis-targeting chimerics (PROTACs), more treatment options may be available for investigation to address dual pathway blockade.

Our study is limited by the proportion (36%) of participants with evaluable samples. Most samples could not be analyzed due to tumor fraction constraints, which were selected based on prior experience demonstrating poor WES performance below these thresholds.²³ Despite this, clinicopathologic characteristics were similar across the cohort, emphasizing the representative nature of the analyzable subset. Our challenge underscores the importance of development of ctDNA sequencing approaches starting from minimal ctDNA fraction and low cfDNA input to allow for analysis of an expanded number of patient samples. Second, though all patients had clinically HER2-positive metastatic disease, a minority of tumors underwent FISH testing, a standard approach in the HER2 testing algorithm.⁷ This limits our ability to correlate our findings from WES with clinical test results. Finally, our cohort is not demographically representative of patients with HER2-positive metastatic breast cancer in the United States and does not include a sufficient number of Black or Hispanic patients. Additional studies should be performed in representative patient populations with special attention to under-represented groups.

In conclusion, we conducted a genomic analysis of patients with HER2-positive MBC receiving TKIs in the metastatic setting via whole exome sequencing of tumor and cfDNA samples. We analyzed a primary cohort of participants with paired samples prior to and after TKI treatment to assess for mechanisms of resistance, followed by validation in a separate cohort of participants with comparable clinicopathologic features. The findings in this study support previously identified pre-clinical resistance mechanisms, such as activating PIK3CA mutations, in a clinical cohort of patients who received anti-HER2 TKIs. Our identification of ESR1 hotspot mutations, may add another layer to the crosstalk between estrogen receptor pathway activation and blockade of HER2 by anti-HER2 therapy. Intriguingly, we further found that mutations in tumor suppressor genes of the WNT pathway were significantly enriched in participants with brain metastases, though this finding will need to be validated in additional cohorts and characterized in functional studies. While we observed many exome-based mutations in oncogenic signaling pathways, several patients remained without a plausible mechanism of resistance. Novel methods in liquid biopsies, such as fragmentomics and epigenomics, may reveal mechanisms of resistance from activation or downregulation of pathways resulting from epigenetic alterations.

Population and Samples

We identified participants with biopsy-proven, metastatic HER2-positive breast cancer, with and without brain metastases, with at least one cfDNA and/or tumor tissue sample collected on DF/HCC IRB-approved protocols 06-213, 09-204, 11–344, 13–198, or 17–318. Inclusion criteria, study design, and outcomes of the 06-213 (lapatinib), 11–344 (neratinib), 13–198 (tucatinib), and 17–318 (trastuzumab, pertuzumab, eribulin) clinical trials have previously been described. Participants provided informed consent for collection of blood, tumor tissue and access to clinical and genetic data for research purposes. Two Streck tubes of blood (20cc) were collected at the time of treatment change, 4–6 weeks after treatment change, and every 3 months if there was no progression and were processed as previously described.^24–27 From this cohort, we retrospectively identified participants who received ≥ 1 anti-HER2 TKI (lapatinib, neratinib and/or tucatinib) for ≥ 6 weeks in the metastatic setting. We included participants with ≥ 1 sample with tumor fraction (TFx) ≥ 9.5% and additional samples with TFx ≥ 4.5%. We identified mutually exclusive cohorts of participants with paired samples with ≥ 6 weeks of tyrosine kinase treatment, participants with samples collected pre-TKI, and participants with samples collected post-TKI. We used previously reported median progression free survival data for lapatinib, neratinib and tucatinib to inform thresholds for acquired and intrinsic resistance.^28–30 Participants who received TKI treatment for 180 days or longer were categorized as having acquired resistance. Participants who received a TKI for less than 180 days were categorized as having intrinsic resistance.

Genomic Analysis

All samples initially underwent ultra-low-pass whole-genome sequencing (ULP-WGS) via the Broad Institute Genomics Platform. Reads were aligned to the GRCh37 reference genome by the BWA-MEM algorithm.^31,32 The Picard Mark Duplicates tool identified duplicate reads.³³ The resulting genomes had an average read depth of 0.2x, with average read lengths of 146 bp. The ichorCNA tool provided estimates of the samples’ TFx from ULP-WGS.²³ The TFx informed all of our subsequent analyses.

The remainder of our genomic analysis relied on whole exome sequencing (WES) data. Whole exomes were sequenced by the Broad Institute Genomics Platform. WES on cfDNA samples employed a Twist bait set; for solid tissue samples an Agilent bait set was used.³⁴ Reads were aligned to the GRCh37 reference using BWA-MEM with Picard Mark Duplicates. This yielded WES with average depths of 20x and average read lengths of 140 bp.

Somatic single nucleotide variants (SNVs) and indels were called via MuTect and Strelka, respectively.^35,36 Filtering steps removed potential false positives arising from FFPE and OxoG artifacts. A panel of normals filter removed potential germline events.³⁷ An additional filter based on BLAT re-alignment removed artifacts stemming from mismapped reads.³⁸ We used OncoKB to label SNVs, indels, and genes with functional annotations.³⁹

Gene-level annotations of oncogene and tumor suppressor gene status were obtained from the literature.⁹ However, two genes in the WNT pathway with mutations in the paired cohort—LZTR1 and CHD8—were missing this annotation. Hence, we annotated these two genes as tumor suppressors based on ancillary sources of evidence: tumor suppressor status annotation in OncoKB (for LZTR1) and prior reports (for CHD8).⁴⁰

We called somatic copy number alterations (CNAs) for each tumor sample in several steps. AllelicCapSeg generated copy number segmentations for each tumor-normal pair of whole exomes.⁴¹ We used ABSOLUTE to compute absolute ploidy for tumors and assign allelic copy numbers to the segmentations.⁴² We then translated ABSOLUTE’s segment-level calls to gene-level copy number annotations. This information allowed us to annotate genes with a variety of CNAs, including gains, amplifications, deep deletions, and biallelic losses. See the Supplemental Methods for details about our criteria for calling these events.

We used PhylogicNDT to infer a clonal phylogeny for each participant from their ABSOLUTE output.⁴³ PhylogicNDT is a Bayesian technique that infers the phylogenetic tree of a tumor from its mutations’ Cancer Cell Fractions (CCFs, i.e., variant allele fractions, adjusted for copy number and tumor purity). PhylogicNDT’s output includes (i) a phylogenetic tree of clones present in the tumor; (ii) an assignment of mutations to those clones; and (iii) the CCF of each clone, in each sample.

For each participant possessing both pre- and post-TKI samples, we labeled each tumor clone as growing, shrinking, or stable based on whether its CCF increased or decreased between pre- and post-TKI samples. See the Supplemental Methods for details of our labeling criteria. Importantly, these three labels cannot be used to explicitly denote treatment sensitivity/resistance; for example, a clone with “shrinking” CCF may still be growing in volume/cell count. However, the labels can give evidence of clones’ relative responses to treatment.

We used MutSig2CV to identify driver genes in our cohort.⁴⁴ MutSig2CV is a statistical model of genes’ background passenger mutation rates, i.e., the genes’ mutation rates under random chance—accounting for gene- and patient-specific attributes. Given the observed mutations in a cohort of tumor samples, MutSig2CV yields a p-value for each gene. We aggregated these gene-level p-values into pathway-level p-values via Fisher’s method, allowing us to identify pathways with significantly more alterations than expected under random chance.⁴⁵ We adjusted these p-values for FDR via Benjamini-Hochberg correction.⁴⁶

Four mutational signatures were identified in our full cohort via SignatureAnalyzer’s implementation of Automatic Relevance Determination Nonnegative Matrix Factorization (ARD-NMF). ARD-NMF is a Bayesian method that finds a parsimonious set of linear factors (signatures) that explain the mutational spectra in a population of tumor samples. These often indicate specific physical processes causing mutations in the tumors.^47,48 See the Supplemental Methods for details about our mutational signature methodology. Three of the four signatures sufficed to explain the mutations in our paired subcohort. These aligned with COSMIC3 SBS signatures for clock-like, capecitabine, and APOBEC mutational processes.⁴⁷

Prior Presentation:

This study was presented at the American Association for Cancer Research meeting, Orlando, FL in April 2023.

Nader-Marta G, Martins-Branco D, de Azambuja E. How we treat patients with metastatic HER2-positive breast cancer. ESMO Open. 2022 Feb;7(1):100343. doi: 10.1016/j.esmoop.2021.100343. Epub 2022 Jan 4. PMID: 34995893; PMCID: PMC8741455.
Rimawi MF, Schiff R, Osborne CK. Targeting HER2 for the treatment of breast cancer. Annu Rev Med. 2015;66:111-28. doi: 10.1146/annurev-med-042513-015127. PMID: 25587647.
Swain, S.M., Shastry, M. & Hamilton, E. Targeting HER2-positive breast cancer: advances and future directions. Nat Rev Drug Discov 22, 101–126 (2023). https://doi.org/10.1038/s41573-022-00579-0
Huang, L., Jiang, S. & Shi, Y. Tyrosine kinase inhibitors for solid tumors in the past 20 years (2001–2020). J Hematol Oncol 13, 143 (2020). https://doi.org/10.1186/s13045-020-00977-0
Samuel W. Brady, Jian Zhang, Daniel Seok, Hai Wang, Dihua Yu; Enhanced PI3K p110α Signaling Confers Acquired Lapatinib Resistance That Can Be Effectively Reversed by a p110α-Selective PI3K Inhibitor. Mol Cancer Ther 1 January 2014; 13 (1): 60–70. https://doi.org/10.1158/1535-7163.MCT-13-0518
Xu X, De Angelis C, Burke KA, Nardone A, Hu H, Qin L, Veeraraghavan J, Sethunath V, Heiser LM, Wang N, Ng CKY, Chen ES, Renwick A, Wang T, Nanda S, Shea M, Mitchell T, Rajendran M, Waters I, Zabransky DJ, Scott KL, Gutierrez C, Nagi C, Geyer FC, Chamness GC, Park BH, Shaw CA, Hilsenbeck SG, Rimawi MF, Gray JW, Weigelt B, Reis-Filho JS, Osborne CK, Schiff R. HER2 Reactivation through Acquisition of the HER2 L755S Mutation as a Mechanism of Acquired Resistance to HER2-targeted Therapy in HER2⁺ Breast Cancer. Clin Cancer Res. 2017 Sep 1;23(17):5123-5134. doi: 10.1158/1078-0432.CCR-16-2191. Epub 2017 May 9. PMID: 28487443; PMCID: PMC5762201.
Wolff AC, Somerfield MR, Dowsett M, Hammond MEH, Hayes DF, McShane LM, Saphner TJ, Spears PA, Allison KH. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: ASCO-College of American Pathologists Guideline Update. J Clin Oncol. 2023 Aug 1;41(22):3867-3872. doi: 10.1200/JCO.22.02864. Epub 2023 Jun 7. PMID: 37284804.
Cocco E, Javier Carmona F, Razavi P, Won HH, Cai Y, Rossi V, Chan C, Cownie J, Soong J, Toska E, Shifman SG, Sarotto I, Savas P, Wick MJ, Papadopoulos KP, Moriarty A, Cutler RE Jr, Avogadri-Connors F, Lalani AS, Bryce RP, Chandarlapaty S, Hyman DM, Solit DB, Boni V, Loi S, Baselga J, Berger MF, Montemurro F, Scaltriti M. Neratinib is effective in breast tumors bearing both amplification and mutation of ERBB2 (HER2). Sci Signal. 2018 Oct 9;11(551):eaat9773. doi: 10.1126/scisignal.aat9773. PMID: 30301790; PMCID: PMC6498841.
Sanchez-Vega F, Mina M, Armenia J, Chatila WK, Luna A, La KC, Dimitriadoy S, Liu DL, Kantheti HS, Saghafinia S, Chakravarty D, Daian F, Gao Q, Bailey MH, Liang WW, Foltz SM, Shmulevich I, Ding L, Heins Z, Ochoa A, Gross B, Gao J, Zhang H, Kundra R, Kandoth C, Bahceci I, Dervishi L, Dogrusoz U, Zhou W, Shen H, Laird PW, Way GP, Greene CS, Liang H, Xiao Y, Wang C, Iavarone A, Berger AH, Bivona TG, Lazar AJ, Hammer GD, Giordano T, Kwong LN, McArthur G, Huang C, Tward AD, Frederick MJ, McCormick F, Meyerson M; Cancer Genome Atlas Research Network; Van Allen EM, Cherniack AD, Ciriello G, Sander C, Schultz N. Oncogenic Signaling Pathways in The Cancer Genome Atlas. Cell. 2018 Apr 5;173(2):321-337.e10. doi: 10.1016/j.cell.2018.03.035. PMID: 29625050; PMCID: PMC6070353.
Lung, D.K., Reese, R.M. & Alarid, E.T. Intrinsic and Extrinsic Factors Governing the Transcriptional Regulation of ESR1. HORM CANC 11, 129–147 (2020). https://doi.org/10.1007/s12672-020-00388-0
Degasperi A, Zou X, Amarante TD, Martinez-Martinez A, Koh GCC, Dias JML, Heskin L, Chmelova L, Rinaldi G, Wang VYW, Nanda AS, Bernstein A, Momen SE, Young J, Perez-Gil D, Memari Y, Badja C, Shooter S, Czarnecki J, Brown MA, Davies HR; Genomics England Research Consortium; Nik-Zainal S. Substitution mutational signatures in whole-genome-sequenced cancers in the UK population. Science. 2022 Apr 22;376(6591):science.abl9283. doi: 10.1126/science.abl9283. PMID: 35949260; PMCID: PMC7613262.
Lee HJ, Park IA, Park SY, Seo AN, Lim B, Chai Y, Song IH, Kim NE, Kim JY, Yu JH, Ahn JH, Gong G. Two histopathologically different diseases: hormone receptor-positive and hormone receptor-negative tumors in HER2-positive breast cancer. Breast Cancer Res Treat. 2014 Jun;145(3):615-23. doi: 10.1007/s10549-014-2983-x. Epub 2014 May 13. PMID: 24820412.
Suda K, Onozato R, Yatabe Y, Mitsudomi T. EGFR T790M mutation: a double role in lung cancer cell survival? J Thorac Oncol. 2009 Jan;4(1):1-4. doi: 10.1097/JTO.0b013e3181913c9f. PMID: 19096299.
Nguyen DX, Chiang AC, Zhang XH, Kim JY, Kris MG, Ladanyi M, Gerald WL, Massagué J. WNT/TCF signaling through LEF1 and HOXB9 mediates lung adenocarcinoma metastasis. Cell. 2009 Jul 10;138(1):51-62. doi: 10.1016/j.cell.2009.04.030. Epub 2009 Jul 2. PMID: 19576624; PMCID: PMC2742946.
Lee Y, Lee JK, Ahn SH, Lee J, Nam DH. WNT signaling in glioblastoma and therapeutic opportunities. Lab Invest. 2016 Feb;96(2):137-50. doi: 10.1038/labinvest.2015.140. Epub 2015 Dec 7. PMID: 26641068.
Koval A, Katanaev VL. Dramatic dysbalancing of the Wnt pathway in breast cancers. Sci Rep. 2018 May 9;8(1):7329. doi: 10.1038/s41598-018-25672-6. PMID: 29743726; PMCID: PMC5943245.
Won HS, Lee KM, Oh JE, Nam EM, Lee KE. Inhibition of β-Catenin to Overcome Endocrine Resistance in Tamoxifen-Resistant Breast Cancer Cell Line. PLoS One. 2016 May 19;11(5):e0155983. doi: 10.1371/journal.pone.0155983. PMID: 27196739; PMCID: PMC4873201.
Brett, J.O., Spring, L.M., Bardia, A. et al. ESR1 mutation as an emerging clinical biomarker in metastatic hormone receptor-positive breast cancer. Breast Cancer Res 23, 85 (2021). https://doi.org/10.1186/s13058-021-01462-3
Jeselsohn R, Bergholz JS, Pun M, Cornwell M, Liu W, Nardone A, Xiao T, Li W, Qiu X, Buchwalter G, Feiglin A, Abell-Hart K, Fei T, Rao P, Long H, Kwiatkowski N, Zhang T, Gray N, Melchers D, Houtman R, Liu XS, Cohen O, Wagle N, Winer EP, Zhao J, Brown M. Allele-Specific Chromatin Recruitment and Therapeutic Vulnerabilities of ESR1 Activating Mutations. Cancer Cell. 2018 Feb 12;33(2):173-186.e5. doi: 10.1016/j.ccell.2018.01.004. PMID: 29438694; PMCID: PMC5813700.
Pegram M, Jackisch C, Johnston SRD. Estrogen/HER2 receptor crosstalk in breast cancer: combination therapies to improve outcomes for patients with hormone receptor-positive/HER2-positive breast cancer. NPJ Breast Cancer. 2023 May 31;9(1):45. doi: 10.1038/s41523-023-00533-2. PMID: 37258523; PMCID: PMC10232442.
Arpino G, Wiechmann L, Osborne CK, Schiff R. Crosstalk between the estrogen receptor and the HER tyrosine kinase receptor family: molecular mechanism and clinical implications for endocrine therapy resistance. Endocr Rev. 2008 Apr;29(2):217-33. doi: 10.1210/er.2006-0045. Epub 2008 Jan 23. PMID: 18216219; PMCID: PMC2528847.
Giuliano M, Trivedi MV, Schiff R. Bidirectional Crosstalk between the Estrogen Receptor and Human Epidermal Growth Factor Receptor 2 Signaling Pathways in Breast Cancer: Molecular Basis and Clinical Implications. Breast Care (Basel). 2013 Aug;8(4):256-62. doi: 10.1159/000354253. PMID: 24415978; PMCID: PMC3808214.
Harold J. Burstein et al., Endocrine Therapy With or Without Inhibition of Epidermal Growth Factor Receptor and Human Epidermal Growth Factor Receptor 2: A Randomized, Double-Blind, Placebo-Controlled Phase III Trial of Fulvestrant With or Without Lapatinib for Postmenopausal Women With Hormone Receptor–Positive Advanced Breast Cancer—CALGB 40302 (Alliance). JCO 32, 3959-3966(2014). DOI:10.1200/JCO.2014.56.7941
Adalsteinsson, V. A. et al. Scalable whole-exome sequencing of cell-free DNA reveals high concordance with metastatic tumors. Nat Commun 8, 1324 (2017).
Parsons HA, Rhoades J, Reed SC, Gydush G, Ram P, Exman P, Xiong K, Lo CC, Li T, Fleharty M, Kirkner GJ, Rotem D, Cohen O, Yu F, Fitarelli-Kiehl M, Leong KW, Hughes ME, Rosenberg SM, Collins LC, Miller KD, Blumenstiel B, Trippa L, Cibulskis C, Neuberg DS, DeFelice M, Freeman SS, Lennon NJ, Wagle N, Ha G, Stover DG, Choudhury AD, Getz G, Winer EP, Meyerson M, Lin NU, Krop I, Love JC, Makrigiorgos GM, Partridge AH, Mayer EL, Golub TR, Adalsteinsson VA. Sensitive Detection of Minimal Residual Disease in Patients Treated for Early-Stage Breast Cancer. Clin Cancer Res. 2020 Jun 1;26(11):2556-2564. doi: 10.1158/1078-0432.CCR-19-3005. Epub 2020 Mar 13. PMID: 32170028; PMCID: PMC7654718.
Lin NU, Guo H, Yap JT, Mayer IA, Falkson CI, Hobday TJ, Dees EC, Richardson AL, Nanda R, Rimawi MF, Ryabin N, Najita JS, Barry WT, Arteaga CL, Wolff AC, Krop IE, Winer EP, Van den Abbeele AD. Phase II Study of Lapatinib in Combination With Trastuzumab in Patients With Human Epidermal Growth Factor Receptor 2-Positive Metastatic Breast Cancer: Clinical Outcomes and Predictive Value of Early [18F]Fluorodeoxyglucose Positron Emission Tomography Imaging (TBCRC 003). J Clin Oncol. 2015 Aug 20;33(24):2623-31. doi: 10.1200/JCO.2014.60.0353. Epub 2015 Jul 13. PMID: 26169615; PMCID: PMC4534525.
Freedman RA, Gelman RS, Anders CK, Melisko ME, Parsons HA, Cropp AM, Silvestri K, Cotter CM, Componeschi KP, Marte JM, Connolly RM, Moy B, Van Poznak CH, Blackwell KL, Puhalla SL, Jankowitz RC, Smith KL, Ibrahim N, Moynihan TJ, O'Sullivan CC, Nangia J, Niravath P, Tung N, Pohlmann PR, Burns R, Rimawi MF, Krop IE, Wolff AC, Winer EP, Lin NU; Translational Breast Cancer Research Consortium. TBCRC 022: A Phase II Trial of Neratinib and Capecitabine for Patients With Human Epidermal Growth Factor Receptor 2-Positive Breast Cancer and Brain Metastases. J Clin Oncol. 2019 May 1;37(13):1081-1089. doi: 10.1200/JCO.18.01511. Epub 2019 Mar 12. PMID: 30860945; PMCID: PMC6494354.
Metzger Filho O, Leone JP, Li T, Tan-Wasielewski Z, Trippa L, Barry WT, Younger J, Lawler E, Walker L, Freedman RA, Tolaney SM, Krop I, Winer EP, Lin NU. Phase I dose-escalation trial of tucatinib in combination with trastuzumab in patients with HER2-positive breast cancer brain metastases. Ann Oncol. 2020 Sep;31(9):1231-1239. doi: 10.1016/j.annonc.2020.05.014. Epub 2020 May 24. PMID: 32461105.
Geyer CE, Forster J, Lindquist D, Chan S, Romieu CG, Pienkowski T, Jagiello-Gruszfeld A, Crown J, Chan A, Kaufman B, Skarlos D, Campone M, Davidson N, Berger M, Oliva C, Rubin SD, Stein S, Cameron D. Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med. 2006 Dec 28;355(26):2733-43. doi: 10.1056/NEJMoa064320. Erratum in: N Engl J Med. 2007 Apr 5;356(14):1487. PMID: 17192538.
Saura C, Oliveira M, Feng YH, Dai MS, Chen SW, Hurvitz SA, Kim SB, Moy B, Delaloge S, Gradishar W, Masuda N, Palacova M, Trudeau ME, Mattson J, Yap YS, Hou MF, De Laurentiis M, Yeh YM, Chang HT, Yau T, Wildiers H, Haley B, Fagnani D, Lu YS, Crown J, Lin J, Takahashi M, Takano T, Yamaguchi M, Fujii T, Yao B, Bebchuk J, Keyvanjah K, Bryce R, Brufsky A; NALA Investigators. Neratinib Plus Capecitabine Versus Lapatinib Plus Capecitabine in HER2-Positive Metastatic Breast Cancer Previously Treated With ≥ 2 HER2-Directed Regimens: Phase III NALA Trial. J Clin Oncol. 2020 Sep 20;38(27):3138-3149. doi: 10.1200/JCO.20.00147. Epub 2020 Jul 17. PMID: 32678716; PMCID: PMC7499616.
Murthy RK, Loi S, Okines A, Paplomata E, Hamilton E, Hurvitz SA, Lin NU, Borges V, Abramson V, Anders C, Bedard PL, Oliveira M, Jakobsen E, Bachelot T, Shachar SS, Müller V, Braga S, Duhoux FP, Greil R, Cameron D, Carey LA, Curigliano G, Gelmon K, Hortobagyi G, Krop I, Loibl S, Pegram M, Slamon D, Palanca-Wessels MC, Walker L, Feng W, Winer EP. Tucatinib, Trastuzumab, and Capecitabine for HER2-Positive Metastatic Breast Cancer. N Engl J Med. 2020 Feb 13;382(7):597-609. doi: 10.1056/NEJMoa1914609. Epub 2019 Dec 11. Erratum in: N Engl J Med. 2020 Feb 6;382(6):586. PMID: 31825569.
6. Church, D. M. et al. Modernizing Reference Genome Assemblies. PLOS Biology 9, e1001091 (2011).
Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. Preprint at https://doi.org/10.48550/arXiv.1303.3997 (2013).
Picard toolkit. Broad Institute, GitHub repository (2019).
Yaldiz, B. et al. Twist exome capture allows for lower average sequence coverage in clinical exome sequencing. Human Genomics 17, 39 (2023).
Cibulskis, K. et al. Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol 31, 213–219 (2013).
Kim, S. et al. Strelka2: fast and accurate calling of germline and somatic variants. Nat Methods 15, 591–594 (2018).
Ellrott, K. et al. Scalable Open Science Approach for Mutation Calling of Tumor Exomes Using Multiple Genomic Pipelines. Cell Syst 6, 271-281.e7 (2018).
Kent, W. J. BLAT—The BLAST-Like Alignment Tool. Genome Res. 12, 656–664 (2002).
Chakravarty, D. et al. OncoKB: A Precision Oncology Knowledge Base. JCO Precis Oncol 1–16 (2017) doi:10.1200/PO.17.00011.
Waks, Z., Weissbrod, O., Carmeli, B. et al. Driver gene classification reveals a substantial overrepresentation of tumor suppressors among very large chromatin-regulating proteins. Sci Rep 6, 38988 (2016). https://doi.org/10.1038/srep38988
Stachler, M. D. et al. Paired exome analysis of Barrett’s esophagus and adenocarcinoma. Nat Genet 47, 1047–1055 (2015).
Carter, S. L. et al. Absolute quantification of somatic DNA alterations in human cancer. Nat Biotechnol 30, 413–421 (2012).
Leshchiner, I. et al. Comprehensive analysis of tumour initiation, spatial and temporal progression under multiple lines of treatment. 508127 Preprint at https://doi.org/10.1101/508127 (2019).
Lawrence, M. S. et al. Discovery and saturation analysis of cancer genes across 21 tumour types. Nature 505, 495–501 (2014).
Fisher, S. R. A. Statistical Methods for Research Workers. (Oliver and Boyd, 1970).
Benjamini, Y. & Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society: Series B (Methodological) 57, 289–300 (1995).
Alexandrov, L. B. et al. The repertoire of mutational signatures in human cancer. Nature 578, 94–101 (2020).
Kasar, S. et al. Whole-genome sequencing reveals activation-induced cytidine deaminase signatures during indolent chronic lymphocytic leukaemia evolution. Nat Commun 6, 8866 (2015).
Istemi Bahceci, Ugur Dogrusoz, Konnor C La, Özgün Babur, Jianjiong Gao, Nikolaus Schultz, PathwayMapper: a collaborative visual web editor for cancer pathways and genomic data, Bioinformatics, Volume 33, Issue 14, July 2017, Pages 2238–2240, https://doi.org/10.1093/bioinformatics/btx149
Seth A. Wander, Ofir Cohen, Xueqian Gong, Gabriela N. Johnson, Jorge E. Buendia-Buendia, Maxwell R. Lloyd, Dewey Kim, Flora Luo, Pingping Mao, Karla Helvie, Kailey J. Kowalski, Utthara Nayar, Adrienne G. Waks, Stephen H. Parsons, Ricardo Martinez, Lacey M. Litchfield, Xiang S. Ye, Chunping Yu, Valerie M. Jansen, John R. Stille, Patricia S. Smith, Gerard J. Oakley, Quincy S. Chu, Gerald Batist, Melissa E. Hughes, Jill D. Kremer, Levi A. Garraway, Eric P. Winer, Sara M. Tolaney, Nancy U. Lin, Sean G. Buchanan, Nikhil Wagle; The Genomic Landscape of Intrinsic and Acquired Resistance to Cyclin-Dependent Kinase 4/6 Inhibitors in Patients with Hormone Receptor–Positive Metastatic Breast Cancer. Cancer Discov 1 August 2020; 10 (8): 1174–1193. https://doi.org/10.1158/2159-8290.CD-19-1390
Gómez Tejeda Zañudo, J., Barroso-Sousa, R., Jain, E. et al. Exemestane plus everolimus and palbociclib in metastatic breast cancer: clinical response and genomic/transcriptomic determinants of resistance in a phase I/II trial. Nat Commun 15, 2446 (2024). https://doi.org/10.1038/s41467-024-45835-6

Yes there is potential Competing Interest. Heather Parsons receives institutional funding from Puma Biotechnology and MERCK. Jose Leone receives research funding from Kazia Therapeutics and consults for Minerva Biotechnologies. Gad Getz receives research funds from IBM and Pharmacyclics, and is also an inventor on patent applications filed by the Broad Institute related to MSMuTect, MSMutSig, POLYSOLVER, SignatureAnalyzer-GPU, and MSIDetect. He is also a founder, consultant, and holds privately held equity in Scorpion Therapeutics.

HER2TKISupplementalTablesFiguresMethods.docx
Dataset 1
SupplementaryTable5HER2TKIManuscript.xlsx
Dataset 2

Download PDF

Version 1

posted

You are reading this latest preprint version

Genomic mechanisms of resistance to tyrosine kinase inhibitors in HER2 amplified breast cancer

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

The genomic landscape of post-TKI HER2-positive advanced breast cancer

Evolution of Clonal Dynamics in TKI resistance

WNT pathway enrichment in HER2-positive brain metastases

DISCUSSION

METHODS

Population and Samples

Genomic Analysis

DECLARATION

REFERENCES

Additional Declarations

Supplementary Files

Status:

Version 1