In premenopausal women

The ASTRRA study (NCT00912548) investigated the efficacy of adding ovarian function suppression (OfS) to tamoxifen (TAM) in 1483 premenopausal HR-positive breast cancer patients who had completed (neo)adjuvant CTx and undergone definitive surgery.⁸Noh WC et al. J Clin Oncol 2018;36(suppl): Abstr 502 With a median follow-up of 63 months, the 5-year estimated DFS was 91.1% in the combination group and 87.5% in the TAM-only group (P=0.029).

The TEXT (NCT00066703) and SOFT (NCT00066690) trials studied the efficacy of OfS in 4891 premenopausal HR-positive HER2-negative breast cancer patients.⁹Regan MM et al. J Clin Oncol 2018;36(suppl): Abstr 503 Participants were randomly assigned to receive EXE plus OfS, TAM+OfS or TAM alone. A composite recurrence risk index (CRI) was calculated for each patient, compromising the patient’s age, nodal status, tumour size and grade, and oestrogen receptor (ER), progesterone receptor (PR), and Ki67 level expression. The overall 8-year distance recurrence-free interval (DRFI) was 91% ranging ~100% for the lowest CRI-level of 0.2 to 63% for the highest CRI level of 3.4. In the combined TEXT and SOFT populations, 4690 patients received OfS. At a median follow-up of 9 years, the 8-year DFS-rate was 86.8% in the EXE+OfS group compared to 82.8% in the TAM+OfS group (HR for recurrence, a second invasive cancer, or death=0.77; 95% CI, 0.67-0.90; P<0.001). Premenopausal patients with HR+/HER2-negative breast cancer and high risk of recurrence by CRI have an approximately 10-15% absolute improvement in the 8-year DRFI when treated with EXE+OfS vs TAM+OfS or TAM alone. The risk in 'lower score' CRI groups was furhter reduced.

Using cyclin-dependent kinase (CDK) 4/6 inhibitors

New data focused on the efficacy and safety of cyclin-dependent kinase (CDK) 4/6 inhibitors. The safety and efficacy of CDK4/6 inhibitor abemaciclib (ABE) was investigated in women with HR-positive, HER2-negative advanced breast cancer (ABC) in the Phase III MONARCH 2 study (NCT02107703).¹⁰Neven P et al. J Clin Oncol 2018;36(suppl): Abstr 1002 During the ASCO 2018, a subgroup analysis of the MONARCH 2 study in pre/perimenopausal women (N=114) was presented. Pre- and perimenopausal women in the MONARCH 2 were required to receive OfS with a gonadotropin-releasing hormone (GnRH) agonist (e.g., goserelin, leuprorelin, triptorelin) before study start. Participants were randomised 2:1 to receive ABE + fulvestrant (F) (N=72) or PBO + F (N=42). The median PFS was not reached for ABE+F and was 10.5 months for PBO+F arm (HR=0.45; 95% CI, 0.26-0.75; P=0.002). The objective response rate (ORR) was 60.8% for ABE+F vs 28.6% for PBO+F (P=0.006). Complete responses (CR) were observed in 3.9% of ABE+F-treated patients vs 0% in patients receiving PBO+F. The most common any-grade adverse events (AEs) for ABE+F were diarrhoea (87.3%), neutropenia (59.2%), and leukopenia (43.7%). In the PBO+F arm, these AEs were observed in 23.8%, 7.1%, and 4.8% of patients respectively.

The Phase III MONALEESA-3 study (NCT02422615) investigated the use of ribociclib (RIB) with F in 726 postmenopausal women with HR-positive, HER2-negative ABC.¹¹Slamon DJ et al. J Clin Oncol 2018;36(suppl): Abstr 1000 The primary efficacy endpoint, investigator-assessed median PFS, was met. Median PFS was 20.5 months for RIB+F vs 12.8 months for PBO+F (HR=0.59; 95% CI, 0.48-0.73; P<0.001). The efficacy was consistent in endocrine treatment-naïve patients and patients pretreated with up to 1 line of endocrine therapy for ABC. The ORR was 41% for RIB+F vs 29% for PBO+F (P=0.003). The clinical benefit rate (CBR) measured 69% vs 60% (P=0.015), respectively. Frequent all-grade AEs that occurred in ≥30% of patients were neutropenia (70% vs 2%), nausea (45% vs 28%), and fatigue (31% vs 33%), all for RIB+F vs PBO+F respectively. Notably, Grade 3/4 neutropenia was observed in 54% (RIB+F) vs 0% (PBO+F) of patients, G3/4 increased ALT in 9% (RIB+F) vs <1% (PBO+F), G3/4 increased AST in 6% (RIB+F) vs 1% (PBO+F), and post-baseline QTcF-interval >480 ms occurred in 6% (RIB+F) vs 3% (PBO+F) of patients.

In the MONALEESA-7 (NCT02278120) subgroup analysis of 672 premenopausal HR-positive, HER2-negative ABC patients, the median PFS was increased when adding RIB to ET (TAM or non-steroidal AI [NSAI])+OfS group vs the PBO plus ET+OfS group.¹²Hurvitz SA et al. J Clin Oncol 2018;36(suppl): Abstr 1047 The improvement in PFS was observed in patients who received one-line of prior CTx (N=94) and patients without prior CTx (N=578). In patients pretreated with CTx, the PFS was 16.6 vs 9.0 months (HR=0.55; 95% CI, 0.31-0.95) and the ORR was 32% vs 27% (P=0.262), both for RIB+ET+OfS vs PBO+ET+OfS respectively. In patients without prior CTx, the PFS was 24.7 vs 14.5 months (HR=0.57; 95% CI, 0.44-0.72) and the ORR was 54% vs 38%, again for RIB+ET+OfS vs PBO+ET+OfS respectively. ORR was also significantly increased in the RIB group without prior CTx. Although adding RIB to ET+OfS was beneficial in premenopausal patients with HR+, HER2- ABC regardless of previous CTx, the data suggests patients who received CTx before entering the MONALEESA-7 study had numerical shorter PFS and lowered ORR when compared to patients without prior CTx.

CDK4/6: pooled and translational data

Four other studies looked at the treatment effect of CDK4/6 inhibition in combination with endocrine therapy (ET).^13-16 A joint analysis of 666 postmenopausal HR+ HER2-negative patients untreated for ABC in the PALOMA 2 study (PAL2 [NCT01740427]) and 521 pre- and postmenopausal HR+ HER2-negative patients whose disease has progressed after prior ET in the PALOMA3 trial (PAL3 [NCT01942135]) showed that adding palbociclib (P) to ET, either L (PAL2) or F (PAL3), had a significant increase in PFS regardless of the length of initial treatment-free interval (TFI [PAL2]) or disease-free interval (DFI [PAL3]).¹³Finn RS et al. J Clin Oncol 2018;36(suppl): Abstr 1023 The analysis also revealed that both Luminal A and Luminal B subtypes benefitted from the addition of PAL to ET in both studies.

The United States Food and Drug Administration (US FDA) pooled analysis on the benefits of CDK4/6 inhibitor in less common breast cancer subsets; PR-negative, de novo metastatic, and lobular histology.¹⁴Gao JJ et al. J Clin Oncol 2018;36(suppl): Abstr 1024 The investigators determined that addition of a CDK4/6 inhibitor to ET provided similar reduction of risk of disease progression or death in less common breast cancer subtypes as is reported for the broad population per product indication. Nevertheless, it was noted that questions remain on the selection of patients who may derive more benefit and in what setting to best prescribe CDK4/6 inhibitors.

The MONALEESA-2 (NCT01958021) investigated the addition of RIB to L vs PBO + L in 668 postmenopausal patients with HR+, HER2-negative ABC.¹⁵Hortobagyi GN et al. J Clin Oncol 2018;36(suppl): Abstr 1022 Gene expression analysis by NanoString 230-gene nCounter® GX Human Cancer Reference panel assessed correlations between high and low messenger RNA expression level and PFS outcomes. The investigators looked at gene expression involved in the breast cancer CDK pathway (ESR1, RB1, E2F1), gene expression implicated in CDK4/6 inhibitor resistance (CDK2, CCNE1, FGFR1), and gene expression involved in alternative pathways (cell cycle control, PI3K pathway, MAPK pathway, RTKs). The addition of RIB was beneficial for all gene expression levels in all genes. However, there was a trend towards greater benefit of RIB in high ESR1 expression and low RTK expression

In the PAL3 study (NCT01942135), mechanisms of resistance to the CDK4/6 inhibitor used were investigated using circulating tumour DNA (ctDNA) analysis in tissue samples of 125 PAL+F patients and 68 PBO+F patients.¹⁶Turner NC et al. J Clin Oncol 2018;36(suppl): Abstr 1001 Evolution of driver gene mutations was more common in patients progressing late on PAL+F treatment but uncommon in patients experiencing early progression. A paired exome analysis was performed in 16 patients receiving PAL+F and showed frequent clonal evolution, but copy number profiles remained constant before and after treatment. Furthermore, the outcomes showed that breast cancer driver mutations are similar regardless of treatment with PAL+F or F alone, with acquired PIK3CA and ESR1 Y537S mutations which may contribute to F resistance. At end-of-treatment (EOT), RB1 mutations emerged in 4.8% of PAL+F patients. RB1 mutations were not observed at EOT in patients receiving PBO+F.

HER2-negative disease

Concluding the ASCO hormone-sensitive breast cancer coverage is the Phase I/II study (NCT02379247) with the novel PI3K inhibitor alpelisib (ALP – BYL719) in combination with nab-paclitaxel (NP) in HER2-negative MBC.¹⁷Sharma P et al. J Clin Oncol 2018;36(suppl): Abstr 1018 The study encompassed a Phase I  3+3 dose-finding part to determine the recommended Phase II dose (RP2D), ORR and PFS, and a Phase II part to expand the ORR and PFS investigations. In total 43 patients enrolled in Phase I (N=10) and Phase II (N=33) and 42 patients were evaluable for response. The ORR with ALP+NP was 57% (CR=2, partial response [PR]=22). Stable disease ≥16 weeks was observed in 21% of patients. The median PFS was 9 months. By next-generation sequencing 17/42 (40%) of patients were found to have tissue and ctDNA PIK3CA mutations. The ctDNA/tissue concordance was 70%. Patients with PIK3CA mutation had better median PFS compared to those without PIK3CA mutations, 13 vs 7 months (HR=0.39; P=0.03). Notable, in the study population 74% had received prior CTx (84% taxane) for MBC, 30% of patients had triple-negative breast cancer (TNBC), and 84% was reported to have visceral metastases. Grade 3/4 AEs were hyperglycaemia (29%), neutropenia (31%), anaemia (12%) and diarrhoea (7%).

Another Phase III study (NCT02253459) study in 405 heavily pretreated anthracycline- and taxane-refractory MBC patients included both TNBC as well as HR-positive, HER2-negative breast cancer patients to investigate the epothilone analogue utidelone (UTI)+C vs C alone.¹⁸Xu B et al. J Clin Oncol 2018;36(suppl): Abstr 1003 The intention-to-treat (ITT) population demonstrated a median OS of 14.8 months in the combination group vs 12.2 months in the C alone group (HR=0.63, 95% CI, 0.45─0.88; P=0.005) and the 24-month OS-rate was 39.0% for UTI+C vs 26.6% C alone (P=0.015). The most common Grade 3 AE in the UTI+C arm was peripheral neuropathy (25%) which was manageable and reversible. UTI was associated with only mild myelosuppression which is remarkable for a microtubule inhibitor. No liver toxicities were reported.

A phase II study (NCT02282345) of neoadjuvant poly(ADP-ribose) polymerase-inhibitor (PARPi) talazoparib (TALA) in operable breast cancer with BRCA mutation (BRCA+) recruited 3 HR-positive, HER2-negative patients and 17 TNBC patients.¹⁹Litton JK et al. J Clin Oncol 2018;36(suppl): Abstr 508 The primary endpoint was residual cancer burden (RCB), with RCB 0 and 1 having a positive expected outcome when compared to higher RCB scores. At the time of publication, 19 patients were evaluable. The investigators reported that once daily TALA given preoperatively without CTx produces significant pCR, 53% (10/19) of patients achieved an RCB0 and 63% (12/19) of patients had an RCB score of 0 or 1. One Grade 4 toxicity was reported (thrombocytopenia [5%]). Grade 3 toxicities were anaemia (40%), neutropenia (15%), and urinary tract infection (5%). Common grade 1/2 toxicities included nausea, fatigue, neutropenia, alopecia, dizziness and dyspnoea. Nine patients continued at a lower dose.

Advances in the management of TNBC

The Phase III BrighTNess (NCT02032277) study investigated the addition of PARPi veliparib (VELI) to CTx in 634 stage II-III TNBC patients and assessed the prognostic and predictive role of homologous recombination deficiency (HRD) assay for carboplatin (CPt) and PARPi response.²⁰Telli ML et al. J Clin Oncol 2018;36(suppl): Abstr 519 Patients received treatment with either VELI+CPt+T, PBO+CPt+T, or dual PBO+T, all followed by cyclophosphamide plus doxorubicin (AC). The HRD-status was determined for 438 patients. The results showed higher rates of pCR in HRD-positive (HRD+) patients across all treatment arms; Patients treated with CPt had the highest pCR in both HRD+ and HRD-negative (HRD-) subsets, regardless of the addition of VELI. Irrespective of the treatment received, HRD+ patients had higher pCR rates when compared to HRD- patients in the same treatment arm.

The Phase II, investigator-initiated, double-blind, randomised-controlled PAKT study (NCT02423603) investigated the addition of novel AKT inhibitor AZD5363 to T as first-line therapy in 140 women with metastatic TNBC (mTNBC).²¹Schmid P et al. J Clin Oncol 2018;36(suppl): Abstr 1007 The primary endpoint was PFS and had a predefined significance level of 0.10. The PFS results in the ITT population showed a median PFS of 5.9 months for AZD5363 vs 4.2 months for PBO (P=0.06 [one-sided]). The median OS was 19.1 months for AZD5363 vs 12.6 months for PBO  (P=0.04 [two-sided]). Activating mutations of the AKT-pathway were found in 24% of patients. These patients with PIK3CA/AKT1/PTEN-alterations had improved PFS outcome (HR=0.30) and longer OS (HR=0.37). Grade 3 and higher AEs included diarrhoea (12% vs 1%), infection (4% vs 1%), neutropenia (3% vs 3%), rash (4% vs 0%) and fatigue (4% vs 0%), all for AZD5363+T vs PBO+T respectively.

The Phase II, double-blind, randomised-controlled LOTUS study (NCT02162719) reported the updated OS results of adding AKT inhibitor ipatasertib (IPAT) to T as first-line therapy in 124 women with mTNBC. Previously, Kim et al. had reported the PFS outcome; median PFS was 6.2 months for the IPAT combination vs 4.9 for PBO+T (stratified HR=0.60; 95% CI, 0.37–0.98; P=0.037)²²Kim SB et al. Lancet Oncol 2017;18(10):1360–72 Patients whose tumour harboured PIK3CA/AKT1/PTEN-alterations had a better outcome when receiving the AKT-inhibitor: 9.0 months vs 4.9 months with PBO (unstratified HR=0.44; 95% CI, 0.20–0.99). The updated OS results presented at ASCO 2018 did not reach statistical significance but showed a trend towards improved 1-year OS-rate in favour of IPAT vs placebo (85% vs 70%).²³Dent R et al. J Clin Oncol 2018;36(suppl): Abstr 1008 The median OS was 23.1 for IPAT+T vs 18.4 months for PBO+T (stratified HR=0.62; 95% CI, 0.37–1.05).

With immunotherapy

A Phase II single-arm study (NCT02730130) assessed the efficacy and safety of programmed death 1 (PD-1) inhibitor pembrolizumab (PEMBRO) in combination with radiotherapy (RTx) in heavily pretreated women (N=17) with mTNBC and at least two measurable lesions of which one requiring RTx. PD-1 ligand (PD-L1) expression status was not required for study participation.²⁴McArthur HL et al. J Clin Oncol 2018;36(suppl): Abstr 1017 The primary endpoint measured an abscopal tumour response (ORR) at Week 13 in the non-irradiated lesion(s). Nine women were evaluable for response. The ORR was 33% (all partial responses: 60%, 54%, and 34% decrease in size) and the disease-control rate (DCR) was 44%. Five patients (56%) had disease progression (PD) at Week 13. Eight women were not evaluable for response, five died due to disease-related complications and 3 progressed before Week 13. The AE-profile was mild, common AEs included fatigue, myalgia and nausea. The authors concluded that the addition of RTx to PD-1 blockade may be a promising strategy for improving response rates in pretreated mTNBC.

The TOPACIO/KEYNOTE-162 (NCT02657889) is a Phase I/II study to evaluate the safety and efficacy of combining PARPi niraparib with PD-1 immune-checkpoint inhibitor PEMBRO in 54 patients with mTNBC.²⁵Vinayak S et al. J Clin Oncol 2018;36(suppl): Abstr 1011 Twelve (22%) of the enrolled TNBC patients had deleterious BRCA mutations (BRCAmut). In 9 (17%) of patients BRCAmut were not tested or showed indeterminate results. The investigators reported 3 (7%) CRs and 10 (22%) PRs matching an ORR of 29%. The DCR was 49% with 9 (20%) patients having stable disease (SD). The remainder 23 (51%) patients had PD. Among the 12 patients with BRCAmut the ORR was 67% (1 CR and 7 PR) and the DCR was 75% (1 SD). Three BRCAmut patients had PD. The median PFS in the BRCAmut group was 8.1 months. PD-L1-positiveness (PD-L1+) was defined as a combined proportion score (CPS) ≥1%. In the PD-L1+ group the ORR was 33% vs 15% in PD-L1-negative patients. Common treatment-related AEs (TRAEs) Grade ≥3 were thrombocytopenia (13%) and anaemia (11%).

Another PD-1 antibody, nivolumab (NIVO), was evaluated as maintenance therapy after five different induction therapy options (RTx or CTx or NIVO single-agent) in 66 metastatic TNBC patients in the adaptive Phase II TONIC trial (NCT02499367).²⁶Kok M et al. J Clin Oncol 2018;36(suppl): Abstr 1012 Among all patients (N=66) the ORR was 20% (2 CRs, 11 PRs). Two patients (3%) had SD >24 weeks, matching a CBR of 23%. Induction with single-agent NIVO resulted in an ORR of 17% (2/12). ORRs for NIVO following induction with other therapies was: RTx 8% (1/12), doxorubicin 35% (6/17), cyclophosphamide 8% (1/12) and cisplatin 23% (3/13).

Finally, the Phase II neoadjuvant GeparNeuvo study (NCT02685059) investigated the addition of PD-L1-inhibitor durvalumab (DURVA) to a taxane-anthracycline-cyclophosphamide (TAC) containing regimen in 174 patients with operable TNBC.²⁷Loibl S et al. J Clin Oncol 2018;36(suppl): Abstr 104 A pCR was observed in 84 of 174 pts (48.3%). Serious AEs (SAEs) were reported in 34.5% of patients and immune-related AEs of special interest (irAESI) were reported in 27.6% of patients. The combination of TAC CTx with DURVA showed a feasible option associated with pCR-rate in TNBC.