Blackwell, Kimberly

PURPOSETissue-based next-generation sequencing (NGS) in metastatic breast cancer (mBC) is limited by the inability to noninvasively track tumor evolution. Cell-free DNA (cfDNA) NGS has made sequential testing feasible; however, the relationship between cfDNA and tissue-based testing in mBC is not well understood. Here, we evaluate concordance between tissue and cfDNA NGS relative to cfDNA sampling frequency in a large, clinically annotated mBC data set.METHODSTempus LENS was used to analyze deidentified records of mBC cases with Tempus xT (tissue) and xF (cfDNA) sequencing results. Then, various metrics of concordance were assessed within overlapping probe regions of the tissue and cfDNA assays (104 genes), focusing on pathogenic variants. Variant allele frequencies of discordant and concordant pathogenic variants were also compared. Analyses were stratified by mBC subtype and time between tests.RESULTSRecords from 300 paired tissue and liquid biopsies were analyzed. Median time between tissue and blood collection was 78.5 days (standard deviation = 642.99). The median number of pathogenic variants/patient was one for cfDNA and two for tissue. Across the cohort, 77.8% of pathogenic tissue variants were found in cfDNA and 75.7% of pathogenic cfDNA variants were found in tissue when tests were ≤ 7 days apart, which decreased to 50.3% and 51.8%, respectively, for > 365 days. Furthermore, the median patient-level variant concordance was 67% when tests were ≤7 days apart and 30%-37% when > 30 days. The median variant allele frequencies of discordant variants were generally lower than those of concordant variants within the same time frame.CONCLUSIONWe observed high concordances between tissue and cfDNA results that generally decreased with longer durations between tests. Thus, cfDNA NGS reliably measures tissue genomics and is likely beneficial for longitudinal monitoring of molecular changes in mBC.

Abstract Genomic testing has transformed clinical care in metastatic breast cancer. Cell-free DNA (cfDNA) has made sequential testing feasible, however, concordance between tissue and cfDNA, sensitivity of cfDNA, and optimal sampling frequency have not been examined on large data sets with clinical annotation. Of the 648 genes in the tissue NGS Tempus xT assay, 105 genes are also included in the Tempus xF cfDNA plasma-based test. Both platforms are run on Illumina and detect SNVs, indels, CNVs, and rearrangements/fusions with high sensitivity and specificity (>90%). This study used the Tempus Labs LENS™ tool to identify the most recent ~3,300 metastatic breast cancer patients who underwent clinical testing and had either tissue-based testing, cfDNA testing or both. Of the 212 patients with xF and xT testing, 87 had xT before xF, 135 had xF before xT, and 7 had the testing performed at the same time. The most common pathogenic germline versus somatic alterations on tissue-based testing and their concordance will also be reported. In summary, this data suggests a variable amount of concordance between tissue-based and plasma-based assays. Timing of assay performance and inclusion of germline DNA might play an important role in contributing to these differences which need further exploration. Table: Count of Reported Variants and xT/xF Concordance by GeneSomatic GenesxF frequency (n)xT frequency(n)Concordance Between xF and xTTP5313911043%PIK3CA848058%ESR1584141%BRCA238217%ERBB2372629%NF1322113%RB1272416%ATM30165%GATA3213441%FGFR13353%PTEN193036%ARID1A251823% Citation Format: Minetta Liu, Jeff Schaffer, Aneta Piwowarczyk, Nike Beaubier, Kimberly Blackwell, Ben Park. Concordance and use patterns of tissue and cell-free based genomic testing in metastatic breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD9-12.

1006 Background: Patients (pts) with ER-/PR- MBC do not benefit from endocrine treatment (tx). Doane et al (Oncogene 2006;25:3994) described a subset of ER-/PR- BC with a gene expression profile similar to ER+ BC but characterized by AR expression and AR-dependent growth in vitro. Hence, we conducted a multicenter phase II trial (NCT00468715) of the AR antagonist, bicalutamide (B) for pts with AR+ ER-/PR- MBC. Methods: Pts with ER-/PR- (IHC <10%) MBC were consented to AR testing, confirmed centrally at MSKCC. If AR+ (DAKO IHC ≥10%), pts were eligible for tx if ECOG performance status (PS) <2 and normal organ function. No limit on prior tx except prior trastuzumab required if HER2+. Eligible AR+ pts who consented to tx received B 150mg orally daily in 28-day cycles (C). Toxicity assessed q4wks, response q12wks. Primary endpoint = clinical benefit rate (CBR): CR + PR + stable disease (ds) >6mo (SD). B would be considered worthy of further study if ≥4/28 pts have clinical benefit. Results: As of 12/21/11,436 pts consented for AR testing. 24 were ineligible, 12 await testing. 47/400 tested were AR+ (12%). 26 received tx with B (6 ineligible for tx, 15 are eligible for tx with B at progression (POD) if clinically appropriate and study slots remains). Two AR+ pts treated with B were ER+ and removed from study. Three have received tx <12wks. Treated pt characteristics (n=24): median(med)age 64 (41-83), PS 0 (0-1), HER2+ 1, visceral metastases 17. Prior chemotherapy: neo/adjuvant 16; med # regimens for MBC 1 (0-8). Med C# 3 (2-49+). Best response: (21 evaluable pts): CBR 19% (95% CI 5-42%), SD>6mo 4, CR/PR 0, SD<6mo 3, POD 14. Conclusions: ~12% of ER-/PR- MBC pts are AR+. For these pts, AR-inhibition with B is feasible, well tolerated, and has activity based on pre-specified criteria. This study has closed to accrual for AR testing as of 1/2012. These results of the primary endpoint are not expected to change by accrual of the 2 remaining pts. Supported by TBCRC/Breast Cancer Alliance/AstraZeneca. [Table: see text]

PURPOSE Evidence-based treatments for metastatic, human epidermal growth factor receptor 2 (HER2)–positive breast cancer to the CNS are limited. We previously reported modest activity of neratinib monotherapy for HER2-positive breast cancer brain metastases. Here we report the results from additional study cohorts. PATIENTS AND METHODS Patients with measurable, progressive, HER2-positive brain metastases (92% after receiving CNS surgery and/or radiotherapy) received neratinib 240 mg orally once per day plus capecitabine 750 mg/m 2 twice per day for 14 days, then 7 days off. Lapatinib-naïve (cohort 3A) and lapatinib-treated (cohort 3B) patients were enrolled. If nine or more of 35 (cohort 3A) or three or more of 25 (cohort 3B) had CNS objective response rates (ORR), the drug combination would be deemed promising. The primary end point was composite CNS ORR in each cohort separately, requiring a reduction of 50% or more in the sum of target CNS lesion volumes without progression of nontarget lesions, new lesions, escalating steroids, progressive neurologic signs or symptoms, or non-CNS progression. RESULTS Forty-nine patients enrolled in cohorts 3A (n = 37) and 3B (n = 12; cohort closed for slow accrual). In cohort 3A, the composite CNS ORR = 49% (95% CI, 32% to 66%), and the CNS ORR in cohort 3B = 33% (95% CI, 10% to 65%). Median progression-free survival was 5.5 and 3.1 months in cohorts 3A and 3B, respectively; median survival was 13.3 and 15.1 months. Diarrhea was the most common grade 3 toxicity (29% in cohorts 3A and 3B). Neratinib plus capecitabine is active against refractory, HER2-positive breast cancer brain metastases, adding additional evidence that the efficacy of HER2-directed therapy in the brain is enhanced by chemotherapy. For optimal tolerance, efforts to minimize diarrhea are warranted.

1005 Background: Evidence-based treatments (tx) for metastatic, HER2+ BCBM are limited. We previously found a central nervous system (CNS) objective response rate (ORR) of 8% (95% CI 2-22%) for the irreversible, EGFR/HER2-targeted kinase inhibitor, neratinib. To enhance CNS activity, we evaluated the combination of neratinib + capecitabine in a subsequent cohort, and report results here. Methods: Pts with measurable BCBM (≥ 1 cm in longest dimension) and no prior lapatinib or capecitabine were eligible. All but 3 had CNS progression after local CNS tx. During 21 day cycles, pts received capecitabine 750 mg/m2 twice daily x 14 days followed by 7 days off + neratinib 240 mg orally once daily. Loperamide prophylaxis (16 mg total daily) was recommended during cycle 1. Brain MRI and non-CNS imaging were repeated every 2 cycles until 18 wks, then every 3 cycles. The primary endpoint was composite CNS ORR, requiring all of the following: ≥50% reduction in volumetric sum of target CNS lesions (central review, VORR), no progression of non-target or non-CNS lesions, no new lesions, no escalating steroids, and no progressive neurologic signs/symptoms. We used a two-stage design with hypotheses ORR 15% and 35% (error rates 5% and 20%), responses in ≥5/19 pts to enter 2ndstage; responses in ≥9/35 [26%] pts to be promising. Results: 39 pts enrolled between 4/2014-11/2016 (2 withdrew before tx, 37 analyzed); median age 51, median prior metastatic lines 2 (range 0-6), 65% had prior WBRT. As of 11/15/16, 23 (62%) patients are alive and 7 remain on protocol tx; median number of cycles initiated = 5 (range 1-26). 18 women (49%) had a VORR (95% CI 32-66%, neurologic exams not yet available on all pts). Overall 12-month survival is 63% (95% CI 43%-77%); 4/7 pts still on protocol therapy have not yet reached 6 cycles. No pts had grade 4 toxicity; 18 (49%) had grade 3 toxicity, with diarrhea most common (32%), and 6 pts discontinued tx for toxicity. Conclusions: The combination of neratinib and capecitabine is active for BCBM with VORR in nearly half of pts, supporting further development of the regimen for BCBM. Updated results will be presented at the meeting. Clinical trial information: NCT01494662.

Abstract Background: HR+ breast cancer in younger vs older patients may have distinct biological features. The Breast Cancer Index (BCI) is a gene expression-based assay that includes two component biomarkers: the HoxB13/IL17BR (H/I ratio), an endocrine response biomarker; and the molecular grade index (MGI), a set of proliferation-related genes. The objective of this study was to assess age-related associations with endocrine sensitivity (H/I), proliferative status (MGI), and quantitative expression of ER and PR. Methods: Data were extracted from the BCI Clinical Database for Correlative Studies, an IRB-approved de-identified database containing clinicopathologic and molecular variables from clinical cases submitted for BCI testing. Molecular results from H/I, MGI, and quantitative (qPCR–based) ER and PR were analyzed across age groups. Chi-squared tests and ANOVA were used to compare the results between age groups (&lt;40y, 40-49y, 50-59y, 60-69y, and ≥70y). Results: Analyses included 19,126 patients (median age at diagnosis 58.6y; 4.5% &lt;40y, 20.6% 40-49y, 28.9% 50-59y, 32.5% 60-69y, and 13.6% ≥70y). Proliferation status (MGI) was significantly higher in patients &lt;40y and 40-49y compared to older groups (P&lt;.0001).H/I analysis indicated a similar distribution of high versus low endocrine responsiveness across all groups (P=.94), except the 40-49y group, in which fewer patients had high H/I (44.2% in &lt;40y, 39.4% in 40-49y, 43.7% in 50-59y, 43.7% in 60-69y, and 43.1% in ≥70y; P=.0001). Median qER increased with age (P&lt;0.0001), while qPR was similar across all age groups except for the 40-49y group (P=.57), in which expression was higher (P&lt;.0001). Conclusion: Results from &gt;19,000 patients with early-stage HR+ breast cancer and BCI testing showed a broad distribution in all variables. Tumors from the youngest patients (&lt;40y) had the highest expression of proliferative genes and the lowest quantitative ER expression. However, endocrine response, according to the H/I biomarker, does not appear to be strongly linked to age. Citation Format: Melisko M, Chien AJ, Poage GM, Salganik M, Schnabel CA, Ruddy KJ, Blackwell K. Gene expression patterns in younger versus older HR+ breast cancer patients: An age-related analysis of HoxB13/IL17BR (H/I), proliferation status, and quantitative hormone receptor expression [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-09-03.

Abstract Introduction: MutHER is a phase II trial that demonstrated the anti-tumor activity of the pan-HER inhibitor neratinib in HER2mut, non-amplified MBC. The major challenges to accrue to this trial were the large number of pts to screen for the 2-3% HER2mut population and the high rate (24%) of poor quality tumor DNA for sequencing. The goals of this ctDNA study were: 1) the concordance of HER2mut detected by ctDNA versus tumor testing; 2) the incidence of ctDNA HER2mut in HER2 non-amplified MBC; 3) changes in HER2mut variant allele frequency (VAF) on neratinib therapy. Methods: A sample size of 30 negative (neg) controls was needed to ensure 90% confidence if ctDNA testing has &gt;90% specificity in detecting HER2mut. Thus, plasma from MBC pts obtained at screening for MutHER trial (Neg control: 40 pts without HER2mut on tumor testing; Positive (pos) control: 14 pts with known HER2mut who received neratinib) were subjected to Guardant360 ctDNA 70-gene panel sequencing (all exons of HER2 included). ctDNA from the 14 neratinib treated pts were also analyzed at week (wk) 4 and upon progression. ctDNA data from MBC pts clinically tested at Guardant Health were interrogated for HER2mut incidence. Results: Among the 14 pts with tumor pos for HER2mut, ctDNA sequencing identified the same HER2mut in 11, discrepant HER2mut in 1, and neg in 2. The 2 pts with ctDNA neg for HER2mut had progressive disease (PD) and stable disease (SD &gt; 6 months) on neratinib, respectively. Among the 40 neg controls, 8 were not evaluable (no detectable ctDNA or assay unsuccessful) and all 32 successfully sequenced cases were neg for HER2mut. The sensitivity and specificity of ctDNA for HER2mut detection was 11/14 (79%, 90% CI: 53-94%) and 32/32 (100%, 90% CI: 91-100%), respectively. Among the 11 paired baseline and wk 4 samples, 9 (82%) had lower HER2mut VAFs at wk 4 than at baseline, with 1 complete response (CR), 1 partial response (PR), 5 SD, and 4 PD at wk 8 as best tumor response. Two pts had higher wk 4 ctDNA HER2mut VAFs and both had radiographic PD at wk 8. The absolute HER2mut VAF levels at wk 4 were significantly associated with TTP (Spearman rho=-0.69, p=0.02) and tumor size change (rho=0.67, p=0.05). The HER2mut VAFs were significantly higher at progression than wk 4 in all pts (p&lt;0.01). One pt acquired a new HER2mut T798I, which is analogous to the gate-keeper mutation EGFR T790M. The incidence of HER2mut without amplification in unselected consecutive MBC clinically tested by Guardant360 was 3% (48/1,584), with mutation pattern similar to published tumor testing data. Conclusions: ctDNA sequencing is sensitive and highly specific in detecting HER2mut, offering a non-invasive method to identify pts for trials of HER2mut-targeted therapy. Decreased HER2mut VAFs at wk 4 was observed in 82% of cases, consistent with the on-target effect of neratinib. Increased HER2mut VAFs at wk 4 is a potential early marker of progression. Citation Format: Cynthia Ma, Ron Bose, Feng Gao, Rachel Freedman, Melinda Telli, Gretchen Kimmick, Eric Winer, Michael Naughton, Matthew Goetz, Christy Russell, Debu Tripathy, Melody Cobleigh, Andres Forero, Timothy Pluard, Carey Anders, Shana Thomas, Jill Anderson, Caroline Bumb, Kimberly Banks, Richard Lanman, Richard Bryce, Alshad Lalani, John Pfeifer, Daniel Hays, Mark Pegram, Kimberly Blackwell, Philippe Bedard, Hussam Al-Kateb, Matthew Ellis. Circulating tumor DNA (ctDNA) sequencing for HER2 mutation (HER2mut) screening and response monitoring to neratinib in metastatic breast cancer (MBC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT011. doi:10.1158/1538-7445.AM2017-CT011

Abstract Purpose: Based on promising preclinical data, we conducted a single-arm phase II trial to assess the clinical benefit rate (CBR) of neratinib, defined as complete/partial response (CR/PR) or stable disease (SD) ≥24 weeks, in HER2mut nonamplified metastatic breast cancer (MBC). Secondary endpoints included progression-free survival (PFS), toxicity, and circulating tumor DNA (ctDNA) HER2mut detection. Experimental Design: Tumor tissue positive for HER2mut was required for eligibility. Neratinib was administered 240 mg daily with prophylactic loperamide. ctDNA sequencing was performed retrospectively for 54 patients (14 positive and 40 negative for tumor HER2mut). Results: Nine of 381 tumors (2.4%) sequenced centrally harbored HER2mut (lobular 7.8% vs. ductal 1.6%; P = 0.026). Thirteen additional HER2mut cases were identified locally. Twenty-one of these 22 HER2mut cases were estrogen receptor positive. Sixteen patients [median age 58 (31–74) years and three (2–10) prior metastatic regimens] received neratinib. The CBR was 31% [90% confidence interval (CI), 13%–55%], including one CR, one PR, and three SD ≥24 weeks. Median PFS was 16 (90% CI, 8–31) weeks. Diarrhea (grade 2, 44%; grade 3, 25%) was the most common adverse event. Baseline ctDNA sequencing identified the same HER2mut in 11 of 14 tumor-positive cases (sensitivity, 79%; 90% CI, 53%–94%) and correctly assigned 32 of 32 informative negative cases (specificity, 100%; 90% CI, 91%–100%). In addition, ctDNA HER2mut variant allele frequency decreased in nine of 11 paired samples at week 4, followed by an increase upon progression. Conclusions: Neratinib is active in HER2mut, nonamplified MBC. ctDNA sequencing offers a noninvasive strategy to identify patients with HER2mut cancers for clinical trial participation. Clin Cancer Res; 23(19); 5687–95. ©2017 AACR.