HER2 Gene Amplification Testing by Fluorescent In Situ Hybridization (FISH): Comparison of the ASCO-College of American Pathologists Guidelines With FISH Scores Used for Enrollment in Breast Cancer International Research Group Clinical Trials

Purpose

ASCO and the College of American Pathologists (ASCO-CAP) recently recommended further changes to the evaluation of human epidermal growth factor receptor 2 gene (HER2) amplification by fluorescent in situ hybridization (FISH). We retrospectively assessed the impact of these new guidelines by using annotated Breast Cancer International Research Group (BCIRG) -005, BCIRG-006, and BCIRG-007 clinical trials data for which we have detailed outcomes.

Patients and Methods

The HER2 FISH status of BCIRG-005/006/007 patients with breast cancers was re-evaluated according to current ASCO-CAP guidelines, which designates five different groups according to HER2 FISH ratio and average HER2 gene copy number per tumor cell: group 1 (in situ hybridization [ISH]–positive): HER2-to-chromosome 17 centromere ratio ≥ 2.0, average HER2 copies ≥ 4.0; group 2 (ISH-positive): ratio ≥ 2.0, copies < 4.0; group 3 (ISH-positive): ratio < 2.0, copies ≥ 6.0; group 4 (ISH-equivocal): ratio < 2.0, copies ≥ 4.0 and < 6.0; and group 5 (ISH-negative): ratio < 2.0, copies < 4.0. We assessed correlations with HER2 protein, clinical outcomes by disease-free survival (DFS) and overall survival (OS) and benefit from trastuzumab therapy (hazard ratio [HR]).

Results

Among 10,468 patients with breast cancers who were successfully screened for trial entry, 40.8% were in ASCO-CAP ISH group 1, 0.7% in group 2; 0.5% in group 3, 4.1% in group 4, and 53.9% in group 5. Distributions were similar in screened compared with accrued subpopulations. Among accrued patients, FISH group 1 breast cancers were strongly correlated with immunohistochemistry 3+ status (P < .0001), whereas groups 2, 3, 4, and 5 were not; however, groups 2, 4 and, 5 were strongly correlated with immunohistochemistry 0/1+ status (all P < .0001), whereas group 3 was not. Among patients accrued to BCIRG-005, group 4 was not associated with significantly worse DFS or OS compared with group 5. Among patients accrued to BCIRG-006, only group 1 showed a significant benefit from trastuzumab therapy (DFS HR, 0.71; 95% CI, 0.60 to 0.83; P < .0001; OS HR, 0.69; 95% CI, 0.55 to 0.85; P = .0006), whereas group 2 did not.

Conclusion

Our findings support the original categorizations of HER2 by FISH status in BCIRG/Translational Research in Oncology trials.

Abstract 136: Phospho-S6 levels correlate with response to Copanlisib (BAY 80-6946) in multiple myeloma

Background: Copanlisib (BAY 80-6946) is a reversible, pan-class I PI3K inhibitor with preferential activity for the alpha isoform, which may be of particular importance in multiple myeloma. Here we demonstrated the in vitro efficacy of copanlisib in a panel of 20 multiple myeloma cell lines. Baseline levels of phospho-S6 (P-S6) correlated with sensitivity to copanlisib, resulting in a potential biomarker of response. In addition, the change of P-S6 post-treatment could be used as a pharmacodynamic biomarker for copanlisib treatment.

Methods: We screened a panel of 20 multiple myeloma cell lines and selected 3 sensitive: NCI-H929, MM.1S, L-363, and 3 resistant: AMO-1, JJN3, COLO-677 for further analysis. We performed apoptosis and cell senescence assays following 72 hours of 50nM and 100nM copanlisib exposure. Cell cycle analysis and induction of apoptosis were performed by FACS after propidium iodide or PI/ANX-V FITC staining, respectively. Cellular senescence was determined by measuring β-galactosidase activity in cells treated for 96 hours. Reverse phase protein array (RPPA) was performed at baseline and post treatment for proteomic analysis with confirmatory western blots. Flow cytometry was also performed to monitor the post-treatment P-S6 level changes.

Results: Copanlisib treatment induced apoptosis in the sensitive cell lines (50-80% AN-V+ cells) but not in resistant cell lines (1-5% AN-V+ cells). An increased cell cycle arrest in G1 was also observed in the sensitive cell lines but not in the resistant lines. The cell senescence assays confirmed apoptosis rather than cell senescence as the mechanism of inhibition of proliferation. RPPA analysis demonstrated lower baseline p-S6 (S235/236, S240/244) protein levels in sensitive compared to resistant cell lines and this was confirmed with western blot analysis. Treatment with copanlisib resulted in a greater decrease in p-S6 in the sensitive cell lines NCI-H929 and L363 (53-83%, 73-93% respectively) than in the resistant cell lines COLO-677 and JJN3 (5-27%, and 38-67%, respectively), which was validated by western blot and phospho-flow. We also showed by RPPA and WB that copanlisib down-regulates pro-survival and proliferation molecules including p-S6K1, p-S6 and p-4EBP1, and upregulates pro-apoptotic PDCD4 in all cell lines, but to a greater extent in sensitive cell lines. Finally, pharmacodynamic p-S6 response remained at different post-treatment time points.

Discussion: A differential response to copanlisib is seen in the myeloma cell line panel. A subgroup of multiple myeloma cell lines demonstrated median IC50 values in the low nanomolar range (5-100nM), and responses correlated with low baseline P-S6. This p-S6 stratified response was only observed with PI3K-alpha inhibitors, but not with inhibitors targeting other PI3K isoforms or pan-PI3K inhibitors. Further studies may allow development of a new patient screening method or companion diagnostic.

Citation Format: Sarah Larson, Mao Yu Peng, Monica Mead, Andrae Vandross, Dylan Conklin, Erika Von Euw, Dennis J. Slamon. Phospho-S6 levels correlate with response to Copanlisib (BAY 80-6946) in multiple myeloma [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 136. doi:10.1158/1538-7445.AM2017-136

A randomized, double-blinded, controlled study of tucatinib (ONT-380) vs. placebo in combination with capecitabine (C) and trastuzumab (Tz) in patients with pretreated HER2+ unresectable locally advanced or metastatic breast carcinoma (mBC) (HER2CLIMB)

TPS1107

Background: Tucatinib (ONT-380) is a highly selective small molecule inhibitor of HER2 kinase with nanomolar potency. Unlike dual HER2/EGFR agents, it does not inhibit EGFR at clinically relevant concentrations, decreasing the potential for EGFR-related toxicities (severe skin rash and diarrhea). In preclinical studies, tucatinib demonstrated synergistic activity with Tz, and was active in HER2+ models of brain metastases (mets). In a Phase 1b study, tucatinib was combined with C and Tz in pts with HER2+ MBC previously treated with trastuzumab emtansine (T-DM1) and Tz. Objective responses were seen, including in pts with brain mets. The combination was well tolerated, with low rates of Gr 3 diarrhea at the recommended dose (300 mg PO BID, equivalent to the single agent MTD). Based on these data, tucatinib is now being evaluated in a study in combination with C and Tz (HER2CLIMB). Methods: The primary study objective is to assess the effect of tucatinib vs. placebo given with C + Tz on progression-free survival (PFS) based on independent central review. Additional objectives include PFS in patients with brain mets, overall survival, ORR, duration of response, clinical benefit rate, and safety. The study population includes adult patients with progressive HER2+ locally advanced or MBC who have had prior treatment with a taxane, Tz, pertuzumab and T-DM1. Patients with brain mets, including untreated or progressive brain mets, may be enrolled. 480 patients will be enrolled in North America, Europe, Israel, and Australia. Patients are receiving C (1000 mg/mg2 PO BID for 14 days of a 21-day cycle) and Tz (6 mg/kg IV once every 21 days), and are being randomized in a 2:1 ratio to tucatinib 300 mg PO BID or placebo. Patients with isolated CNS progression may continue on study treatment after undergoing local CNS therapy. An independent Data Monitoring Committee is monitoring patient safety. Clinical trial information: NCT02614794.

Predictors of prolonged benefit from palbociclib (PAL) plus fulvestrant (F) in women with endocrine-resistant hormone receptor–positive/human epidermal growth factor receptor 2–negative (HR+/HER2–) advanced breast cancer (ABC) in PALOMA-3

1050

Background: PAL+F improved progression-free survival (PFS) over F + placebo (P) in patients (pts) with endocrine-resistant HR+/HER2– ABC. We examined factors predictive of long-term benefit on PAL+F. Methods: Pre/postmenopausal pts with HR+/HER2– ABC that progressed on prior endocrine therapy (ET) were randomized 2:1 to PAL (125 mg/d oral [3 wk on, 1 wk off]) + F (500 mg) or P+F. Characteristics of pts with prolonged benefit (treatment [tx] duration ≥18 mo for PAL+F; ≥12 mo for P+F based on median PFS and tx duration) were compared with the intent-to-treat (ITT) population. Results: PAL+F improved PFS vs P+F (11.2 vs 4.6 mo; hazard ratio, 0.50). By Aug 2016, 138 pts had long-term benefit: 100/347 (29%) pts on PAL+F received tx for ≥18 mo, including 70 (20%) who received > 2 y (26–39 cycles). In contrast, 38/174 (22%) pts on P+F received ≥12 mo of tx; only 16 (9%) received > 2 y (27–38 cycles). No apparent differences in baseline characteristics of pts with long-term benefit were observed between groups except that a greater proportion of those on P+F had a single site of disease involvement (40% PAL+F vs 63% P+F). Pts with long-term benefit on PAL+F had lower rates of visceral disease (42% vs 60%), liver metastases (18% vs 40%), and ≥3 disease sites (27% vs 39%) at baseline vs the ITT population; no difference in sensitivity to prior ET was observed (84% vs 79%). Objective response rate (ORR) was higher among pts with prolonged benefit on PAL+F vs ITT (36% vs 26%). Conclusions: PAL+F is associated with prolonged benefit in about a third of pts treated with the combination in PALOMA-3. These pts achieve higher ORR compared to other study pts and the benefit is independent of baseline site and number of metastatic recurrences and prior endocrine sensitivity. Benefit from F alone is less prolonged and appears limited to those with 1 site of disease involvement. The analysis confirms the efficacy of PAL+F in HR+ ABC with visceral recurrence. Biomarker analyses are ongoing in pts with long-term benefit to understand molecular features predictive of tx sensitivity. Funding: Pfizer. Clinical trial information: NCT01942135.

Overall survival results from the randomized phase II study of palbociclib (P) in combination with letrozole (L) vs letrozole alone for frontline treatment of ER+/HER2– advanced breast cancer (PALOMA-1; TRIO-18)

1001

Background: Preclinical data identified a synergistic role for P and hormone blockade in blocking growth of ER+ breast cancer (BC) cell lines. PALOMA-1 was an open-label phase II trial comparing progression-free survival (PFS) in patients (pts) with advanced ER+/HER2– BC treated with P+L or L alone. Median PFS increased with addition of P to L to 20.2 mos (vs 10.2 mos with L alone; HR = 0.488), with an acceptable safety profile, leading to accelerated approval by the US FDA. These results were confirmed in the phase 3 PALOMA-2 trial. At the time of the final PFS analysis, overall survival (OS) data were immature with only 61 events in both arms and a median follow-up of < 30 mos with a trend in favor of P+L vs L (37.5 vs 33.3 mos; HR = 0.813; P= 0.211). Here we present final OS results. Methods: PALOMA-1 was a 2-part study evaluating P+L in ER+/HER2– advanced BC. Part 1 enrolled postmenopausal pts with this subtype using only ER+/HER2– while Part 2 enrolled pts of this subtype additionally screened for CCND1 amplification and/or loss of p16. The primary endpoint was investigator-assessed PFS. Secondary endpoints included objective response rate, OS, safety, and correlative biomarker studies. A total of 165 pts were randomized; 66 in Part 1 and 99 in Part 2. Baseline characteristics were balanced between treatment arms. In both parts, pts were randomized 1:1 to receive P+L or L alone. OS data were collected as well as post-study therapy. Results: As of Dec 2016, there were 116 OS events. Median OS was 37.5 mos (95% CI: 31.4, 47.8) with P+L vs 34.5 mos (95% CI: 27.4, 42.6) for L (HR = 0.897 [95% CI: 0.623, 1.294]; P= 0.281). Median OS was 37.5 vs 33.3 mos (HR = 0.837; P= 0.280) for Part 1 and 35.1 vs 35.7 mos (HR = 0.935; P= 0.388) for Part 2. 78.6% of pts in the P+L arm received post-study systemic therapy vs 86.4% in the L arm. More pts in the L arm received ≥3 lines of therapy (37% vs 18%). Further subgroup analyses and details on post-study therapies will be presented. Conclusions: In PALOMA-1, P+L provided a statistically non-significant trend towards an improvement in OS. Survival data from the phase III, PALOMA-2 study is awaited. Sponsor: Pfizer; Clinical trial information: NCT00721409.

Proteasome ubiquitin receptor PSMD4 is an amplification target in breast cancer and may predict sensitivity to PARPi

Poly (ADP-ribose) polymerase 1 (PARP1) is an enzyme involved in DNA repair under investigation as a chemotherapeutic target. Current randomized phase three trials of PARPi in metastatic breast cancer are limited to patients with documented BRCA1/2 mutations and no biomarker of PARPi beyond BRCA status is available. In an effort to identify novel biomarkers for PARP inhibition, we created a cell line (HCC1187/TALRES) resistant to the PARP1 inhibitor talazoparib. Herein we show by array-CGH that HCC1187/TALRES has a selective loss of the proteasome ubiquitin receptor PSMD4 amplicon resulting in significant down-regulation of PSMD4. Conversely, we find that breast cancer cell lines that have copy number gain or amplification for PSMD4 are significantly more sensitive to talazoparib. Functional studies reveal that knock-down of PSMD4 in amplified breast cancer cells and loss of the PSMD4 amplicon result in knock-down of PARP1 protein. We show that PSMD4 is amplified and overexpressed in breast cancer and its overexpression correlates with poor survival. Knock-down of PSMD4 results in a significant decrease in cell growth. We provide evidence that PSMD4 is a proteasomal amplification target in breast cancer that PSMD4 amplification confers sensitivity to PARP inhibition, and that PSMD4 amplification is lost in the process of acquiring resistance to PARPi. Finally, this study shows not only that PSMD4 copy number correlates with PARPi sensitivity, but also, that it may be a better predictor of sensitivity to PARPi than BRCA1/2 mutation.

Effect of dendritic cells (DC) transduced with chimeric antigen receptor (CAR) on CAR T-cell cytotoxicity

144

Background: T cells interacting DC could be superior in T cell cytotoxicity. CD141+/Cleg9a+ intra-tumoral DC play a critical role in tumor cytotoxicity. Therefore, combining intra-tumoral DC in CAR T cell would safely increase localized CAR T cell cytotoxicity. We hypothesized that bioengineered DC compartment could be an excellent source for enhanced CAR T cell cytotoxicity. Methods: DC precursors and T cells of PBMC were transduced with a CAR (pCCL-anti-CD33-4-1BB-CD3z-T2A-GFP; CAR-DC or CAR T). For comparison, additional DC were transduced with 4-1BB cDNA (pCCL-4-1BB-T2A-GFP; 4-1BB-DC) or mock control (pCCL-eGFP). In addition to lentivirus transduction, differentiation of DC in vitro employed Flt3L/GM-CSF/IL-4. Transduced CAR T and CAR-DC were sorted by GFP expression at day 5. After further 10 days of culture, cells were harvested and analyzed for phenotype. An acute myeloid leukemia (AML) cell line (Kasumi-1) was treated with CAR T +/- CAR-DC, 4-1BB-DC, or mock control for functional assays. Results: Frequencies of cells expressing CD141+/Cleg9a+ were higher in 4-1BB-DC vs. control DC (33% vs. 1.5%). After mixing CAR T and CAR-DC (5X105) with Kasumi-1 (1X105) for 6 hours, CAR T/CAR-DC showed 100% Kasumi-1 cell cytotoxicity compared to 70% of CAR T by trypan blue. CAR T/CAR-DC also demonstrated higher Annexin V positive Kasumi-1 cells compared with CAR-T (91% vs. 52%). CAR T with or without CAR-DC were also assessed with multiplex immunoassays. CAR T cells mixed with CAR-DC induced higher level of IFN-gamma (10,316 vs. 6,186 pg/ml), IL-2 (68,840 vs. 64,708 pg/ml), and TNFalpha (1,361 vs. 905 pg/ml) (Kasumi-1 cells mixed with CAR-T cells of 10 E/T ratio) than CAR T cells. CAR-DC produced significantly higher IL-12 cytokine production (1,352 vs. 161 pg/ml) than CAR T cells in response to CD33 but independent to T cells, confirmed by comparing IL-12 production with CAR T/4-1BB-DC. Conclusions: These data show that in vitro differentiation of DC bearing 4-1BB increases CD141+/Cleg9a+ DC population and that interaction with CAR-DC to CAR T cells enhances anti-AML cytotoxicity. Our finding may implicate the development of CAR-DC therapy combined to CAR T cells to increase the efficiency of cancer immunotherapy.

A chemically-modified miR-21 inhibitor (ADM-21) as a novel potential therapy in bladder cancer

335

Background: MicroRNAs are key regulators of cancer signaling pathways.Targeting their expression has therapeutic potential. MicroRNA-21 (miR-21) acts as an oncogene in several cancers. Here we describe its functional significance in bladder oncogenesis and the therapeutic properties of a novel chemically-modified miR-21 inhibitor (ADM-21). Methods: MiR-21 expression was assessed by RT-PCR in a panel of human bladder cancer tissues and cell lines. Transcriptomic profiling of 28 bladder cancer cell lines performed to identify miR-21-dependent gene signatures. ADM-21 is a chemically-modified (phosphorothioate backbone and locked-nuclei-acid) antisense oligo against miR-21 with high potency and stability.The efficacy of ADM-21 treatment was evaluated in vitro and in vivo. Results: MiR-21 was up-regulated by 5-fold in bladder tumors relatively to normal tissue and by 3-fold in advanced compared to early stage bladder cancers. The 28 cancer cell lines were stratified in 3 groups (high, intermediate, low) according to miR-21 levels. Transcriptomic analysis revealed a 15-gene signature that negatively correlated with miR-21 levels. Protein Phosphatase 2 Regulatory Subunit B Isoform A (PPP2R2A) displayed tumor suppressive properties in bladder cancer. Moreover, PPP2R2A represents a novel miR-21 direct target gene and a negative regulator of the AKT/mTOR pathway. ADM-21 (10uM) treatment suppressed 43.5% the growth and 90% the invasiveness of 5367 bladder cancer cells. Also, intravenous (I.V.) administration of ADM-21 was proven more effective than intraperitoneal (I.P.) administration in mouse xenografts. Specifically, 15mg/kg of ADM-21 (I.V.) every 5 days for 3 cycles reduced 37% and 47% (day 16) the growth rate of tumors originating from injections of 5367 and RT-112 cells, respectively. Conclusions: Integrative analysis of human bladder tumors and cell lines revealed a novel 15-gene signature that correlates with miR-21 levels. PPP2R2A is a miR-21 direct target and regulator of the AKT/mTOR pathway. ADM-21effectively reduces bladder cancer growth in vitro and in vivo. Upon completion of ADM-21 toxicology studies and confirm safety we will design phase I clinical trial for advanced bladder cancer.

Abstract P3-04-15: The PI3K-inhibitor, copanlisib, has selective activity in luminal breast cancer cell lines and shows robust combined activity with hormonal blockade and CDK-4/6 inhibition in ER+ breast cancer cell line xenografts

Background: Genetic and epigenetic alterations in the PI3K/mTOR and cyclin D:CDK-4/6:Rb signaling axes occur frequently in breast cancer and have been attributed to resistance to both ER- and HER2-directed therapeutics. Pharmacologically targeting CDK-4/6 in combination with hormonal blockade provides clinical benefit in patients with advanced ER+ breast cancer. In this study, we evaluated the activity of the pan-class I PI3K inhibitor, copanlisib (BAY-80-6946), with potent alpha and delta activity as a single agent or in combination with CDK-4/6 inhibition and hormonal blockade in a panel of breast cancer cell lines.

Methods: The growth inhibitory activity of copanlisib was evaluated against a large panel of 48 breast cancer cell lines molecularly characterized by genomic, transcriptomic and proteomic profiling. IC50 values were determined from direct cell counts using a Z1-particle counter. The activity of copanlisib in combination with hormone blockade and CDK-4/6 inhibition, by palbociclib, was assessed in two cell line xenograft models of ER+ breast cancer; MCF7(PIK3CA-E545K) and ZR751(PIK3CA WT). For xenograft studies, tumor bearing mice were treated once weekly (BID) by intravenous injection with clinically achievable doses of copanlisib (10 mg/kg) as single agent or in combination with tamoxifen or fulvestrant with or without 75 mg/kg daily palbociclib for 21 days.

Results: A broad range of IC50 values (0.491-895 nM), with a high degree of separation between sensitive and resistant histologically defined subgroups were determined for copanlisib, indicating the potential for a wide therapeutic window. Luminal subtype, the presence of activating mutations in PIK3CA, high levels of ER, HER2, HER3 and EGFR protein enriched for sensitivity to copanlisib. Activating mutations of KRAS and BRAF were associated with resistance to copanlisib. Single agent copanlisib induced significant tumor growth inhibition (TGI) relative to vehicle control in each of the xenograft models. Modest increases in anti-tumor activity were achieved when copanlisib was combined with hormonal blockade by either tamoxifen or fulvestrant. However, robust tumor regressions were observed with the triple combinations of copanlisib-palbociclib-tamoxifen and copanlisib-palbociclib-fulvestrant. Furthermore, these triple combinations achieved a statistically significant improvement in anti-tumor activity over the standard of care combination of palbociclib plus fulvestrant. Each of the single agent and treatment combinations tested were well tolerated in animals.

Discussion: These preclinical data illustrate the potent and selective activity of the pan class I PI3K inhibitor copanlisib in luminal breast cancers and support the clinical investigation of copanlisib in combination with CDK-4/6 inhibition and hormonal blockade in ER+ breast cancer.

Citation Format: O'Brien NA, Conklin D, Luo T, Ayala R, Issakhanian S, Kalous O, Von Euw E, Politz O, Wilhelm S, Childs BH, Hurvitz SA, Slamon DJ. The PI3K-inhibitor, copanlisib, has selective activity in luminal breast cancer cell lines and shows robust combined activity with hormonal blockade and CDK-4/6 inhibition in ER+ breast cancer cell line xenografts [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-04-15.

Abstract P1-12-07: Phase Ib/II single-arm trial evaluating the combination of everolimus, lapatinib and capecitabine for the treatment of patients with HER2-positive metastatic breast cancer with progression in the CNS after trastuzumab (TRIO-US B-09)

Background: Improving outcomes for patients with HER2+ CNS metastases remains an unmet clinical need. Lapatinib (L) plus capecitabine (C) yields a 20% objective response rate (ORR) in the CNS in patients with previously treated HER2+ breast cancer brain metastases (Lin N, Clin Cancer Res 2009). Everolimus (E), an oral inhibitor of the mammalian target of rapamycin (mTOR), penetrates into the CNS in murine xenograft models (Meikle L, J Neurosci 2008). TRIO-US B09 is an investigator-initiated trial evaluating the safety and clinical activity of the novel combination of L+C+E for the treatment of patients with HER2+ breast cancer brain metastases.

Methods: Patients with trastuzumab-pretreated, HER2+ metastatic breast cancer (MBC) with progression of disease (PD) in the brain and a measurable brain lesion participated. Patients were excluded if they had a prior mTOR inhibitor or an ECOG PS&gt;2. Prior L and/or C, and prior surgery and/or radiation to the brain were allowed. The primary endpoint was CNS ORR at 12 weeks (cycle 3) by RECIST 1.1. Secondary endpoints included safety, progression-free survival, overall survival and extra-CNS ORR. To test the safety of the combination of L+C+E, a 3+3 dose escalation phase was conducted (starting doses: L 1000 mg QD, E 5 mg QD, C: 750 mg/m2 BID d1-14). Treatment was given Q21 days. Patients were evaluated for dose limiting toxicities during C1. Tumor imaging was conducted every 3 cycles. MRI of the brain was performed every 2 cycles through cycle 6 and then every 3 cycles. Neurological symptom assessment was conducted on day 1 of every cycle. Study participants continued to receive treatment until PD, unacceptable toxicity or withdrawal of consent for 12 mos.

Results: Nineteen patients were enrolled at 11 sites in the US and treated with at least one dose of study drug. Of 18 patients with data available, median age was 58.5 (45-68), median number of systemic therapies for MBC was 2 (0-6), and 94.4% had prior radiation and/or surgical resection of brain metastases. 10 patients participated in the dose escalation phase of the study. The maximum tolerated doses were determined to be L 1000 mg QD, E 10 mg QD + C 1000 mg/m2BID days 1-14; however, given tolerability concerns, dose expansion proceeded with Cohort 2 dose for C (750 mg/m2 BID d1-14). Of 17 eligible patients with imaging results available to date, 2 (12%) had a partial response in the CNS at week 12, one of whom continues on study (currently in cycle 13). Stable disease was observed in 7 patients. The most common grade 3/4 adverse events (AE) (CTCAE v4.0) related to E and/or L in 18 treated patients were anorexia (5.5%), dehydration (5.5%), diarrhea (17%), fatigue (5.5%), fever (5.5%) hyperglycemia (5.5%), hypokalemia (11%), and oral mucositis (17%).

Conclusions: This is the first report of this regimen for patients with HER2+ MBC to the brain. This regimen is generally well-tolerated and shows promising activity in the CNS of heavily pretreated patients. Final efficacy and toxicity analyses for all 19 patients will be presented.

Citation Format: Hurvitz SA, Martinez DA, Singh R, Taguchi J, Chan D, Dichmann R, Castrellon A, Barstis J, Hu E, Berkowitz J, Mani A, DiCarlo B, Smalberg I, Hobbs E, Slamon DJ. Phase Ib/II single-arm trial evaluating the combination of everolimus, lapatinib and capecitabine for the treatment of patients with HER2-positive metastatic breast cancer with progression in the CNS after trastuzumab (TRIO-US B-09) [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-12-07.

Abstract OT2-01-12: ENCORE 602: A randomized, placebo-controlled, double-blind, multicenter phase 2 study (with a phase 1b lead-in) of atezolizumab with or without entinostat in patients with advanced triple negative breast cancer (aTNBC)

Background: Atezolizumab, a humanized anti-PDL1 antibody, has shown encouraging single agent activity in triple negative breast cancer. Entinostat is an oral, class I selective histone deacetylase (HDAC) inhibitor. In animal models, entinostat has been shown to selectively reduce immunosuppressive myeloid derived suppressor cells (MDSCs) and regulatory T cells (Tregs), enhancing response to immune checkpoint blockade. It is hypothesized that entinostat in combination with atezolizumab will show improved efficacy compared to atezolizumab alone.

Trial Design: ENCORE 602 is a Phase 1b/2 study evaluating the combination of entinostat plus atezolizumab in patients with aTNBC. The study has 2 phases: an open-label Dose Determination Phase (Phase 1b) followed by Phase 2. The objective of the Dose Determination Phase is to establish the recommended Phase 2 dose (RP2D) of weekly entinostat when given in combination with atezolizumab 1200 mg every 3 weeks. Phase 2 will evaluate the efficacy and safety of entinostat at the RP2D with atezolizumab in patients with aTNBC in a randomized (1:1), double-blind, placebo-controlled setting. The randomization will be stratified by geographic location (US vs ex-US).

Key Eligibility Criteria: Eligible patients will have 1) histologically- or cytologically-confirmed triple negative breast carcinoma that is either metastatic (stage IV of the TNM classification) or locally recurrent and not amenable to local curative treatment, 2) measurable disease based on imaging studies within 28 days before the first dose of study drug, and 3) received 1-2 prior lines of systemic therapy for locally recurrent and/or metastatic disease. Previous treatment with a PD-1/PD-L1-blocking antibody or a HDAC inhibitor is not permitted.

Specific Aims: In Phase 2, the primary endpoint is progression free survival (PFS), as assessed by the investigators using RECIST 1.1. Secondary endpoints include PFS by immune response RECIST (irRECIST), overall response rate, clinical benefit rate, overall survival, safety, and duration and time to response for those patients achieving a complete or partial response. Exploratory endpoints include PK, protein lysine acetylation, and immune correlates.

Statistical Methods: The primary analysis of PFS will be performed using a stratified log-rank test. Estimation of the hazard ratio for treatment effect will be determined using a stratified Cox proportional hazards model. 60 PFS events are estimated to provide 80% power to detect the targeted improvement in PFS with one-sided significance level of 0.1. An independent data safety monitoring board will meet at regular intervals to oversee trial conduct and patient safety.

Accrual: Up to 88 evaluable patients are anticipated if the study completes all phases of evaluation (6-18 patients in Phase 1b, 70 patients in Phase 2). The study was activated in May 2016 (NCT02708680).

Citation Format: Forero A, Stroyakovskiy D, Cha E, Cruickshank S, Hasapidis J, Meyers ML, Slamon DJ. ENCORE 602: A randomized, placebo-controlled, double-blind, multicenter phase 2 study (with a phase 1b lead-in) of atezolizumab with or without entinostat in patients with advanced triple negative breast cancer (aTNBC) [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr OT2-01-12.

Palbociclib and Letrozole in Advanced Breast Cancer

BACKGROUND

A phase 2 study showed that progression-free survival was longer with palbociclib plus letrozole than with letrozole alone in the initial treatment of postmenopausal women with estrogen-receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative advanced breast cancer. We performed a phase 3 study that was designed to confirm and expand the efficacy and safety data for palbociclib plus letrozole for this indication.

METHODS

In this double-blind study, we randomly assigned, in a 2:1 ratio, 666 postmenopausal women with ER-positive, HER2-negative breast cancer, who had not had prior treatment for advanced disease, to receive palbociclib plus letrozole or placebo plus letrozole. The primary end point was progression-free survival, as assessed by the investigators; secondary end points were overall survival, objective response, clinical benefit response, patient-reported outcomes, pharmacokinetic effects, and safety.

RESULTS

The median progression-free survival was 24.8 months (95% confidence interval [CI], 22.1 to not estimable) in the palbociclib-letrozole group, as compared with 14.5 months (95% CI, 12.9 to 17.1) in the placebo-letrozole group (hazard ratio for disease progression or death, 0.58; 95% CI, 0.46 to 0.72; P<0.001). The most common grade 3 or 4 adverse events were neutropenia (occurring in 66.4% of the patients in the palbociclib-letrozole group vs. 1.4% in the placebo-letrozole group), leukopenia (24.8% vs. 0%), anemia (5.4% vs. 1.8%), and fatigue (1.8% vs. 0.5%). Febrile neutropenia was reported in 1.8% of patients in the palbociclib-letrozole group and in none of the patients in the placebo-letrozole group. Permanent discontinuation of any study treatment as a result of adverse events occurred in 43 patients (9.7%) in the palbociclib-letrozole group and in 13 patients (5.9%) in the placebo-letrozole group.

CONCLUSIONS

Among patients with previously untreated ER-positive, HER2-negative advanced breast cancer, palbociclib combined with letrozole resulted in significantly longer progression-free survival than that with letrozole alone, although the rates of myelotoxic effects were higher with palbociclib-letrozole. (Funded by Pfizer; PALOMA-2 ClinicalTrials.gov number, NCT01740427 .).

Abstract CT011: Safety and efficacy results from a phase I dose-escalation trial of the PARP inhibitor talazoparib in combination with either temozolomide or irinotecan in patients with advanced malignancies

Background: Talazoparib is the most potent PARP inhibitor in clinical development (IC50 = 0.57 nMol/L). When talazoparib is combined with DNA damaging agents a synergistic effect mediated through enhanced PARP inhibition and PARP trapping is seen. Here we report first in human combination studies of talazoparib with temozolomide (TEM) or irinotecan (IRI) with biomarker correlates of response. Methods: A phase I dose escalation study evaluated escalating doses of talazoparib (? 0.5 mg PO QD) with either TEM (? 25 mg/m2 PO days 1-5) (Arm A) or IRI (? 25 mg/m2 IV q2 weeks) (Arm B) every 28 days in patients with advanced malignancies. In both arms the dose of talazoparib was escalated first up to the single agent maximum tolerated dose (MTD) of 1.0 mg before the dose of chemotherapy was increased. The primary endpoint was the determination of the maximum tolerated dose (MTD), and secondary endpoints included pharmacokinetics, tumor response, and biomarkers. DLTs were assessed during the 1st cycle of each dose level. RECIST version 1.1 assessment was done every 8 weeks. Prior TEM or IRI was permitted. A subset of patients had tumor assessment using next generation sequencing (NGS) to identify genomic aberrations distinct from BRCA mutations. Results: 41 patients received escalating doses (0.5-1.0 mg) of talazoparib and either TEM or IRI: 19 patients in Arm A and 22 in Arm B. Median age was 57 (21-77), all PS were 0 or 1 and the median number of prior chemotherapy regimens was 6 (1-15). The MTD for each arm was talazoparib 1.0 mg and 37.5 mg/m2 for either TEM or IRI. Confirmed partial responses (PR) were seen in 4/7 (57%) germline BRCA wild-type platinum-resistant ovarian cancer patients, and PRs were seen in one patient each with Ewing's Sarcoma, cervical adenocarcinoma, small cell lung cancer and triple negative breast cancer. There was a correlation between response and the presence of deleterious somatic mutations in non-BRCA DNA repair genes (e.g. PALB2, RAD51D, MSH2). In the ovarian patients, Homologous Recombination Deficiency (HRD) scores ?42 was correlated with response. The most common grade 3/4 AE's (?5%) related to the treatment combination talazoparib + TEM were neutropenia (28%), anemia (33%) and thrombocytopenia (33%), and for the combination talazoparib + IRI were thrombocytopenia (13%), anemia (27%) and neutropenia (31%). No significant PK interactions were seen between talazoparib and either TEM or IRI. Conclusions: The combination of talazoparib with either TEM or IRI is generally well tolerated in patients with heavily pre-treated advanced malignancies. Regardless of histology, there was a correlation with the presence of specific genomic alterations (not confined to DNA repair) and a PR by RECIST. These data support further evaluation of talazoparib in combination with TEM or IRI in tumors to evaluate efficacy and safety with a focus on relevant somatic mutations, pathway predictors and/or response/resistance biomarkers.

Citation Format: Zev A. Wainberg, J. Randolph Hecht, Gottfried E. Konecny, Jonathan W. Goldman, Saeed Sadeghi, Bartosz Chmielowski, Arun Singh, Richard S. Finn, Diego Martinez, Lisa Yonemoto, John Glaspy, Dennis J. Slamon. Safety and efficacy results from a phase I dose-escalation trial of the PARP inhibitor talazoparib in combination with either temozolomide or irinotecan in patients with advanced malignancies. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr CT011.

Abstract 2828: Preclinical activity of abemaciclib as a single agent or in combination with anti-mitotic or targeted therapies for breast cancer

The cyclin D:CDK-4/6:Rb axis is dysregulated in a number of different cancers and is implicated in resistance to hormonal therapy in breast cancer. Pharmacologically targeting cyclin dependent kinase 4 and 6 (CDK4 and 6) has proven to be a successful therapeutic approach in ER+ breast cancer (BC). This study aimed to identify the molecular subtypes of BC that are sensitive to the novel CDK4 and 6 inhibitor, abemaciclib, and identify the best combination strategies for the clinical development.

Growth inhibition activity of abemaciclib was assessed in a panel of 46 BC cell lines molecularly characterized by genomic, transcriptomic and proteomic profiling. IC50 values were determined from direct cell counts using a Z1-particle counter. In vivo activity of abemaciclib was assessed in cell line xenograft models of ER+ and HER2+/ER+ BCs. For ER+ BC, mice were treated daily with clinically achievable doses of abemaciclib (50-75 mg/kg) as a single agent or in combination with tamoxifen or fulvestrant. Combinations with trastuzumab, docetaxel and tamoxifen were assessed in the HER2+/ER+ xenografts.

Sensitivity to abemaciclib was observed predominately in multiple luminal BC cell lines and a small subset of triple negative cell lines that had intact Rb signaling. Activating mutations in PIK3CA also marked for abemaciclib sensitivity. Abemaciclib potentiated the anti-proliferative effects of cytotoxic/anti-mitotic agents when given simultaneously or 48 hours prior to treatment in vitro. Significant tumor growth inhibition (TGI) was observed with single agent abemaciclib in the ER+ BC cell line xenografts. In ZR751 xenografts, the addition of either tamoxifen or fulvestrant to abemaciclib induced complete inhibition of tumor growth for the 12 weeks of treatment. In the MCF7 model, treatment was withdrawn after five weeks, which triggered tumor regrowth in each of the single agent arms. However, complete responses were maintained in the combination arms for a further six weeks post drug withdrawal. In HER2 amplified xenografts, abemaciclib single agent treatment induced significant TGI in trastuzumab sensitive and resistant xenografts, and combination with trastuzumab further increased this anti-tumor effect. The addition of tamoxifen to this combination induced a further increment in TGI. Consistent with the in vitro findings, the combination of abemaciclib and the anti-mitotic agent docetaxel was not antagonistic in vivo, and the addition of docetaxel to the triple combination of abemaciclib, trastuzumab, and tamoxifen induced the most efficacy of any of the treatment arms tested. Combinations were well tolerated in animals.

These data highlight the potential of abemaciclib to have single agent activity in addition to combined activity with anti-mitotic or targeted therapies for breast cancer.

Citation Format: Neil A. O’Brien, Dylan Conklin, Tong Luo, Ondrej Kalous, Erika von Euw, Sara A. Hurvitz, Richard P. Beckmann, Colleen Mockbee, Dennis J. Slamon. Preclinical activity of abemaciclib as a single agent or in combination with anti-mitotic or targeted therapies for breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2828.

Abstract 1206: Preclinical evaluation of targeting Notch-3 in breast cancer

Introduction: Notch-3 overexpression has been implicated in the development of breast cancer (BC) and is associated with poor outcomes. A critical challenge to eliminating treatment resistance in breast cancer likely relates to the presence of cancer stem cells (CSCs) that maintain the ability to differentiate and divide indefinitely. We postulate that targeted eradication of CSCs is possible using a Notch3 antibody drug conjugate (ADC) without irreversibly reducing stem cell viability in vital normal tissues. PF-06650808, is an ADC comprised of a humanized anti-Notch-3 antibody linked to an auristatin-based cytotoxic agent. To better understand the therapeutic index of targeting Notch-3, we evaluated PF-06650808 across a large panel of BC lines and normal cells and correlated response with Notch-3 levels. PF-06650808 was also evaluated in a murine BC xenograft model.

Methods: Response to PF-06650808 and control ADC was evaluated across a panel of BC and normal cell lines by a 2D proliferation assay. Notch-3 mRNA expression was measured by flow cytometry (FC) and RPPA. MDA-MB-468 (triple negative BC, TNBC) tumor bearing mice were randomized into 4 arms of 8 mice and treated with 3 mg/kg PF-06650808 or control-ADC (days 0, 4, 8 & 12), 10 mg/kg docetaxel (q week) or vehicle control.

Results: High expression of Notch-3 was detected in multiple BC cell lines by RPPA and FC. BC cell lines with elevated levels of Notch-3 were sensitive to PF-06650808 (HCC1187, MDA-MB-468, HCC1143, HCC70, EFM-19, HCC202). Responders were also enriched for TNBC. All normal cell lines were resistant to PF-06650808 ADC. When treated with a control ADC against a non-relevant target, all cell lines exhibited IC50s between 5-50ug/ml, indicating that the sub-0.5ug/ml responses seen with the Notch-3 ADC were target-dependent. Durable complete tumor regressions were observed in PF-06650808-treated mice bearing MDA-MB-468 TNBC cell line xenografts.

Conclusions: Sensitivity to a novel anti-Notch-3-ADC is associated with high expression of Notch-3 in BC cell lines. Normal cells are resistant to PF-06650808, possibly predicting a better therapeutic index than seen with other Notch inhibitors. Xenograft studies evaluating the in vivo efficacy of PF-06650808 in a panel of xenografts with varying levels of Notch-3 expression will be presented. Ongoing experiments are exploring the potency of PF-06650808 on CSCs. Our data will help identify breast cancer subtypes most likely to respond to a Notch3-ADC based on high tumor/normal target concentration as well as its effects on CSCs.

Citation Format: Sara A. Hurvitz, Erika von Euw, Neil O’Brien, Dylan Conklin, Chuhong Hu, Jiaying Zhuo, Alice Zhao, Frank Calzone, Hsiao-Wang Chen, Judy Dering, Ken Geles, Puja Sapra, Dennis J. Slamon. Preclinical evaluation of targeting Notch-3 in breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1206.

Abstract CT066: First-in-human phase I trial of the oral PLK4 inhibitor CFI-400945 in patients with advanced solid tumors

Background: CFI-400945 is a first-in-class, potent, selective, orally active inhibitor of Polo-like kinase 4 (PLK4) (Ki = 0.26 nM), a conserved master regulator of centriole duplication that is important for maintenance of genomic integrity. In preclinical studies, CFI-400945 demonstrated robust anti-tumor activity, including durable complete tumor regression, in a large number of patient-derived xenograft models from multiple tumor types (breast, ovarian, pancreas, prostate carcinomas; some derived from heavily pre-treated patients). The objectives of this first-in-human phase I trial are to establish safety, tolerability, pharmacokinetics, and recommended phase II dose (RP2D) of CFI-400945.

Methods: Patients with advanced solid tumors age &gt;18 years, with ECOG performance status of 0-1, adequate organ function and RECIST-measurable disease progressing on standard treatments are eligible. Dose escalation follows a standard 3+3 design, with a starting dose of 3mg once daily continuous oral dosing based upon preclinical data and a severely toxic dose to 10 percent (STD10) in rats of 3 mg/kg. The primary endpoint is the incidence of dose limiting toxicities (DLTs) during the first cycle. Safety assessments using CTCAE version 4.03 criteria are performed weekly during the first three cycles and then every two weeks. Response assessments are performed every two cycles.

Results: From April/14 to December/15, 31 patients were enrolled across eight pre-defined dose levels (3, 6, 11, 16, 24, 32, 48 and 72 mg). No DLT events have been observed. Dose escalation at 96mg is currently ongoing. The most frequent treatment-related adverse events (trAEs) include fatigue (24%), diarrhea (17.2%), nausea (17.2%), decreased appetite (13.8%) and vomiting (6.9%). All trAEs were grade 1 or grade 2. Fifteen serious adverse events (SAEs) have occurred in 9 patients, all considered unrelated to CFI-400945 treatment. Preliminary PK results estimate a half-life of about 10 hours, with Cmax and AUC showing dose proportionality. Two patients enrolled at the 48 mg dose level have completed &gt;6 cycles, including a patient with KRAS mutant colorectal cancer who achieved 24% reduction in target lesions and &gt;50% reduction in serum CEA levels.

Conclusions: CFI-400945 is well tolerated at doses up to 72mg with a favorable PK profile. Preliminary evidence of anti-tumor activity has been observed. Exploration of 96mg daily dosing is ongoing and once the RP2D has been established exploration of anti-tumor activity in biomarker-driven expansion arms of CFI-400945 in indications including advanced breast cancer is planned. Updated results of this ongoing trial will be presented at the meeting.

Citation Format: Philippe L. Bedard, David W. Cescon, Graham Fletcher, Trish Denny, Richard Brokx, Peter Sampson, Mark R. Bray, Dennis J. Slamon, Tak W. Mak, Zev A. Wainberg. First-in-human phase I trial of the oral PLK4 inhibitor CFI-400945 in patients with advanced solid tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr CT066.

Abstract 2736: Regulation of cytokinesis by polo-like kinase 4

Introduction. The serine/threonine mitotic kinase, polo-like kinase 4 (PLK4), is known to play a critical role in centrosome duplication in preparation for cell division. Based on our preliminary observations with a PLK4 kinase inhibitor, we investigated the possibility that PLK4 may also play a role in regulation of cytokinesis.

Experimental Procedures. Immunofluorescence was used to localize PLK4 and phospho-PLK4 in cultured human breast, ovarian, and colorectal cancer cell lines throughout the cell cycle without and with CFI-400945 PLK4 inhibitor and MG-115 protease inhibitor. Flow cytometry and videomicroscopy were used to analyze the consequences of PLK4 inhibition on cytokinesis.

Results. Using immunofluorescence, PLK4 was localized to centrosomes; however, we also found that phospho-PLK4 was cleaved and distributed to kinetochores (metaphase and anaphase), cleavage furrow (telophase), and middle body (cytokinesis) during cell division in colorectal, ovarian, and breast cancer cells. Distribution of phospho-PLK4 to the cleavage furrow and middle body raised the possibility that this kinase plays a functional role in cytokinesis. Using either CFI-400945 PLK4 kinase inhibitor or the MG-115 protease inhibitor, we found that PLK4 accumulated in centrosomes with inhibition of translocation of PLK4 to the middle body. This change in subcellular distribution of PLK4 was associated with generation of large, multi-nucleated tumor cells or tumor cells with polyploidy. Videomicroscopy confirmed that treatment with CFI-400945 PLK4 inhibitor was associated with prevention of cellular abscission in treated cells.

Conclusions. These observations demonstrate a role for phospho-PLK4 in facilitation of cytokinesis. A regulatory role for PLK4 in cytokinesis makes it a potential target for therapeutic intervention in appropriately selected cancers.

Citation Format: Michael F. Press, Bin Xie, Simon Davenport, Yu Zhou, Neil O’Brien, Michael Palazzolo, Tak Mak, Joan Brugge, Dennis J. Slamon. Regulation of cytokinesis by polo-like kinase 4. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2736.

Efficacy and safety of palbociclib in combination with letrozole as first-line treatment of ER-positive, HER2-negative, advanced breast cancer: expanded analyses of subgroups from the randomized pivotal trial PALOMA-1/TRIO-18

Background

Palbociclib is an oral small-molecule inhibitor of cyclin-dependent kinases 4 and 6. In the randomized, open-label, phase II PALOMA-1/TRIO-18 trial, palbociclib in combination with letrozole improved progression-free survival (PFS) compared with letrozole alone as first-line treatment of estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative, advanced breast cancer (20.2 months versus 10.2 months; hazard ratio (HR) = 0.488, 95 % confidence interval (CI) 0.319–0.748; one-sided p = 0.0004). Grade 3–4 neutropenia was the most common adverse event (AE) in the palbociclib + letrozole arm. We now present efficacy and safety analyses based on several specific patient and tumor characteristics, and present in detail the clinical patterns of neutropenia observed in the palbociclib + letrozole arm of the overall safety population.

Methods

Postmenopausal women (n = 165) with ER+, HER2-negative, advanced breast cancer who had not received any systemic treatment for their advanced disease were randomized 1:1 to receive either palbociclib in combination with letrozole or letrozole alone. Treatment continued until disease progression, unacceptable toxicity, consent withdrawal, or death. The primary endpoint was PFS. We now analyze the difference in PFS for the treatment populations by subgroups, including age, histological type, history of prior neoadjuvant/adjuvant systemic treatment, and sites of distant metastasis, using the Kaplan-Meier method. HR and 95 % CI are derived from a Cox proportional hazards regression model.

Results

A clinically meaningful improvement in median PFS and clinical benefit response (CBR) rate was seen with palbociclib + letrozole in every subgroup evaluated. Grade 3–4 neutropenia was the most common AE with palbociclib + letrozole in all subgroups. Analysis of the frequency of neutropenia by grade during the first six cycles of treatment showed that there was a downward trend in Grade 3–4 neutropenia over time. Among those who experienced Grade 3–4 neutropenia, 71.7 % had no overlapping infections of any grade and none had overlapping Grade 3–4 infections.

Conclusion

The magnitude of clinical benefit seen with the addition of palbociclib to letrozole in improving both median PFS and CBR rate is consistent in nearly all subgroups analyzed, and consistent with that seen in the overall study population. The safety profile of the combination treatment in all subgroups was also comparable to that in the overall safety population of the study.

Electronic supplementary material

The online version of this article (doi:10.1186/s13058-016-0721-5) contains supplementary material, which is available to authorized users.