Key words advanced breast cancer - metastases - treatment - mutation testing - immunotherapy - PARP - CDK4/6 -
BRCA1/2
- PD-L1
Introduction
In the last few years, a few targeted therapies have been developed for patients with metastatic breast cancer and subsequently introduced into clinical practice [1 ], [2 ], [3 ], [4 ]. Some of these therapies, such as the CDK4/6 inhibitors or pertuzumab and T-DM1, had already been developed for subgroups of patients that could be identified by immunohistochemistry. Other drugs, such as the PI3K inhibitor alpelisib or the PARP inhibitors, require testing for genetic changes in the tumour or in the germline in addition for the indication. This focused approach to ensure that patients are selected for whom effectiveness will be particularly high and that patients who do not have this characteristic are spared the side effects is the optimal procedure when introducing new drugs. This review article summarises the latest developments in advanced breast cancer
and in the area of biomarkers based on the latest presentations (e.g., ASCO, AACR 2020) and publications.
Treatment of Patients with Advanced HER2-Positive Breast Cancer
Treatment of Patients with Advanced HER2-Positive Breast Cancer
Effective treatment of cerebral metastases
The treatment of patients with HER2-positive advanced breast cancer is characterised by the use of targeted anti-HER2 drugs [5 ]. Trastuzumab, pertuzumab and T-DM1 are very effective anti-HER2 drugs used in the treatment of HER2-positive breast cancer. The tyrosine kinase inhibitors lapatinib and neratinib also have their place. Just in the past year, the tyrosine kinase inhibitor tucatinib was presented through the prospective randomised phase III study HER2CLIMB [6 ]. In patients with advanced breast cancer, who all had pretreatment with trastuzumab, pertuzumab and T-DM1, a comparison between therapy with trastuzumab + capecitabine and tucatinib + trastuzumab + capecitabine showed that progression-free survival and overall survival were improved. A new analysis has now been presented, which focussed on the 291 patients who already had cerebral metastases at the start of the study. In these patients, a
clinically significant advantage was seen for the triple combination. In patients who were treated with tucatinib + trastuzumab + capecitabine, the median CNS-related progression-free survival was prolonged from 4.2 to 9.9 months. The hazard ratio was 0.32 (95% CI: 0.22 – 0.48) [7 ]. The median overall survival was prolonged from 12.0 to 18.1 months. The hazard ratio was 0.58 (79% CI: 0.40 – 0.85). Similar differences between the treatment arms were also seen in groups of patients who had either active or stable brain metastases. It can be concluded that the addition of tucatinib to treatment with capecitabine and trastuzumab is also associated with a marked improvement of the clinical treatment situation in patients with cerebral metastases.
High activity in the area of treatment development for HER2-positive patients
Other anti-HER2 drugs, which are currently being developed for clinical use, include, for example, pyrotinib, margetuximab and trastuzumab-deruxtecan.
Slightly improved progression-free survival was shown for margetuximab, which is an anti-HER2 antibody optimised for ADCC (antibody-dependent cell-mediated cytotoxicity) [8 ]. In the case of this antibody, the ADCC is dependent on a germline variant of genes CD16 and CD32; patients who will benefit most from such therapy can possibly be identified with this.
An improvement in progression-free survival was found for treatment with pyrotinib and capecitabine compared with treatment with lapatinib and capecitabine [9 ]. This was confirmed in the PHOEBE phase III study [10 ]. In the study, after pretreatment with trastuzumab and taxanes, patients were treated either with lapatinib and capecitabine or with pyrotinib and capecitabine. The median progression-free survival was prolonged from 6.8 to 12.5 months (hazard ratio: 0.39; 95% CI: 0.56).
Many studies are currently being conducted with the antibody-drug conjugate (ADC) trastuzumab-deruxtecan. In a single-arm early phase I/II study, this anti-HER2 ADC produced a median progression-free survival of 14.8 months (95% CI: 13.8 – 16.9), even after several previous treatment lines [11 ]. This drug could also be effective in patients with low expression of HER2 (score 1+ or 2+). It is therefore being tested not only in HER2-positive patients (e.g., in the post-neoadjuvant therapy setting, in the metastatic situation after T-DM1 and in comparison with T-DM1, and others), but also in patients with an expression of HER2 that suffices to bind the antibody but without classic overexpression of HER2.
Treatment of Patients with Advanced TNBC or BRCA-Associated Breast Cancer
Treatment of Patients with Advanced TNBC or BRCA-Associated Breast Cancer
Immunotherapies
Since atezolizumab in combination with nab-paclitaxel is already licensed in the first treatment line for patients with metastatic breast cancer, the results of the KEYNOTE-355 study, which tests the PD-1 antibody pembrolizumab also in the first treatment line, have also been published [45 ].
The KEYNOTE-355 study included patients with advanced TNBC who had not yet had any pretreatment in the metastatic stage. The study compared a combination of chemotherapy and pembrolizumab with chemotherapy alone. The permitted chemotherapy included nab-paclitaxel, paclitaxel or gemcitabine/carboplatin. The study design is shown in [Fig. 1 ]. The primary study aim was the progression-free survival, for which a hierarchical procedure was chosen with regard to PD-L1 testing (for a summary of PD-L1 testing see [12 ]). This means that the population that has a CPS score ≥ 10 is tested primarily and if significance is reached, the greater population with a CPS score of ≥ 1 was tested. A total of 847 patients were randomised, of whom 566 were to be treated with pembrolizumab + chemotherapy and 281 with chemotherapy alone (2 : 1 randomisation).
Fig. 1 KEYNOTE-355 study design. a Pembrolizumab 200 mg intravenous (i. v.) every 3 weeks (Q3W); b Chemotherapy dosing regimens are as follows: nab-paclitaxel 100 mg/m2 i. v. on days 1, 8, and 15 every 28 days, paclitaxel 90 mg/m2 i. v. on days 1, 8, and 15 every 28 days, gemcitabine 1000 mg/m2 /carboplatin AUC 2 on days 1 and 8 every 21 days; c Normal saline; d Treatment may be continued until confirmation of progressive disease; e Based on RECIST v 1.1 assessed by a central imaging vendor; f PD-L1 assessed at a central laboratory using PD-L1 ICH 22C3 pharmDx assay and measured using the combined positive score (CPS; number of PD-L1-positive tumour cells, lymphocytes, and macrophages divided by total number of tumour cells × 100); g To be presented at a later date. CNS: central nervous system; ECOG: Eastern Cooperative Oncology Group; PD-L1:
programmed death ligand 1; R: randomized; TNBC: triple-negative breast cancer.(Modified after: [45 ]
)
Of the 847 patients, 323 had a CPS score of ≥ 10 (38.1%) and 636 patients had a CPS score > 1% (75.1%).
The primary study aim was reached when the CPS ≥ 10 tumours were analysed. In this TNBC cohort the addition of pembrolizumab to the chemotherapy prolonged the median PFS from 5.6 to 9.7 months. The hazard ratio was 0.65 (95% CI: 0.49 – 0.86, p = 0.0012). The population with CPS ≥ 1 could therefore be examined. Here the median PFS was 5.6 vs. 7.6 months and the hazard ratio was 0.74 (95% CI: 0.61 – 0.90, p = 0.0014). Since the required p-value in these multiple consecutively planned tests was 0.0011, this result was not formally statistically significant. The Kaplan-Meier curves of these analyses are shown in [Fig. 2 ] and [3 ]. Analyses of overall survival are not yet available. There was no new information about the side effect profile. Thus, data are now available showing that pembrolizumab can be combined with different chemotherapy drugs in the KEYNOTE-355 treatment situation.
Fig. 2 Progression-free survival in the KEYNOTE-355 study in the population of patients with a tumour CPS score of ≥ 10 (primary analysis). a Prespecified p-value boundary of 0.00411 not met. Hazard ratio (CI) analysed based on a Cox regression model with treatment as a covariate stratified by the randomization stratification factors. Data cut-off December 11, 2019. HR: hazard ratio; CI: confidence interval.(Modified after: [45 ]
)
Fig. 3 Progression-free survival in the KEYNOTE-355 study in the population of patients with a tumour and CPS score of ≥ 1. a Prespecified p-value boundary of 0.00411 not met. Hazard ratio (CI) analysed based on a Cox regression model with treatment as a covariate stratified by the randomization stratification factors. Data cut-off December 11, 2019. HR: hazard ratio; CI: confidence interval.(Modified after: [45 ]
)
Chemotherapy as a trigger of an immune response
For some chemotherapy agents and certain dosing schedules it is already known that modulation of the immune system can take place [13 ]. Eribulin and vinorelbine are two of these drugs for which it is suspected that they can have an immunological effect in this way. In a preclinical study, for instance, triple-negative breast cancer cell lines were incubated together with immune cells of the innate immune defence system with eribulin, vinorelbine, docetaxel, paclitaxel and ixabepilone and the immune reaction was measured using the gene expression profile of cytokines and immune checkpoint genes. This showed that eribulin and vinorelbine but not the taxanes can stimulate interferon expression in the immune cells [14 ].
Against this background, combinations of checkpoint inhibitors with these chemotherapy drugs would be of particular interest. In the phase-I/II ENHANCE1 study a combination of eribulin and pembrolizumab was tested for toxicity and efficacy [15 ]. This study included 167 patients with advanced TNBC with 0 – 2 previous treatments. No new unexpected toxicities were seen in addition to those that were expected because of the eribulin therapy or pembrolizumab therapy. A response rate of 25.8% was seen in patients in the first treatment line and in 21.8% in higher treatment lines [15 ]. These response rates are promising and thus this combination could be further developed in further phase III studies.
PARP inhibitors in patients with HRD markers other than BRCA1 and BRCA2
The PARP inhibitors olaparib and talazoparib are licensed for the treatment of HER2-negative patients with advanced breast cancer if a germline BRCA1/2 mutation has been found [16 ], [17 ], which is the case in ca. 10% of cases of TNBC and in ca. 4 – 5% of HER2-negative HR-positive patients [18 ], [19 ], [20 ], [21 ], [22 ]. For veliparib, which can be combined with chemotherapy in a nearly regular dose, it was shown that the addition of veliparib to chemotherapy with carboplatin and paclitaxel can improve the prognosis of patients with HER2-negative advanced breast cancer if a BRCA1/2 germline mutation had been found [23 ] (BROCADE study).
A small phase II study has now been reported (SWOG S1416), which treated HER2-negative patients either with veliparib and cisplatin or with cisplatin alone. Patients were included if they were treated in the first or second treatment line and could be assigned to one of the following groups [46 ].
Group 1 (germline BRCA group): germline mutation in BRCA1/2
Group 2 (BRCA-like group): tumour mutation in BRCA1/2, germline mutation in one of 36 DNA repair genes (BROCA-HR), BRCA1 promoter hypermethylation, myChoice score of ≥ 42
Group 3 (BRCA-unlike): patients could not be assigned to either of the other two groups.
The study design is shown in [Fig. 4 ]. The addition of veliparib did not confer any benefit in the group of BRCA-unlike patients, and also not in the group of patients with a BRCA1/2 germline mutation. The absence of a difference in the group of patients with the BRCA1/2 germline mutation was initially surprising, but it must be considered that this group of 37 patients was small and that olaparib and talazoparib had been licensed during the study for the treatment of patients with a BRCA1/2 germline mutation. In the BRCA-like group (group 2), however, a statistically significant difference was seen with a median progression-free survival of 4.2 months in patients treated with cisplatin and 5.9 months in patients who had been treated with cisplatin + veliparib (hazard ratio: 0.53; 95% CI: 0.34 – 0.83). This shows that apart from the group of patients with a germline mutation in BRCA1/2 there are other BRCA-like characteristics (HRD
score, mutations in other genes, BRCA1 hypermethylation) that could define efficacy of PARP inhibitors.
Fig. 4 SWOG-S1416 study design. a TNBC defined as oestrogen receptor (ER) and progesterone receptor (PgR) immunohistochemical (ICH) nuclear staining of ≤ 1% and HER2 negative per ASCO/CAP guidelines; b Randomization stratified by number of prior cytotoxic regimens for metastatic disease (0 vs. 1). BID: bis in die; MBC: metastatic breast cancer; TNBC: triple-negative breast cancer. Modified after: [46 ].
The TBCRC-048 study (Olaparib Extended Study) is another small phase II study, which addressed a similar question [24 ]. Patients were included in this study who had a germline or tumour mutation in one of the following genes: ATM, ATR, BARD1, BRIP1, CHEK2, FANCA, FANCC, FANCD2, FANCE, FANCF, FANCM, MRE11A, NBN, PALB2, PTEN, RAD50, RAD51C, RAD51D and other unnamed genes or a somatic tumour mutation in BRCA1/2. The patients were treated with olaparib monotherapy. Most germline mutations (87% of the patients) were found in ATM, CHEK2, PALB2 or tumour mutations in BRCA1/2 . The response rate was 33% in the group of patients who had a germline mutation outside BRCA1/2. It was interesting that all 11 patients in whom a germline PALB2 mutation was present benefited from the olaparib therapy (ORR 82%, CBR 100%). In patients with a tumour mutation in BRCA1 or BRCA2 31% had a response to the olaparib therapy [24 ].
Treatment of Patients with Advanced HER2-Negative, Hormone Receptor-Positive Breast Cancer
Treatment of Patients with Advanced HER2-Negative, Hormone Receptor-Positive Breast Cancer
Aromatase inhibitor or fulvestrant as combination partner in the hormone-sensitive situation
The therapy sequence in patients with advanced HER2-negative hormone receptor-positive breast cancer is becoming ever more important given the increasing treatment options [25 ], [26 ], [27 ]. Not only the sequence in the metastatic situation is of importance but also the way anti-hormone therapies are used in the adjuvant setting. Treatment with CDK4/6 inhibitors has become established as standard in the first treatment line of advanced breast cancer. The efficacy of this treatment has been confirmed both in combination with aromatase inhibitors and with fulvestrant. The choice of endocrine combination partner is particularly important, however, because treatments after the CDK4/6 inhibitors could possibly use fulvestrant as combination partner. The efficacy of the PI3K inhibitor alpelisib was already established some time ago in the SOLAR-1 study. In this study, a combination
of fulvestrant with alpelisib was compared with fulvestrant therapy alone [28 ]. Progression-free survival was prolonged from 5.7 months to 11.0 months in patients with PIK3CA mutations. The combination partner fulvestrant could thus be of benefit in the second therapy line after CDK4/6 inhibitors for possible treatment with the PI3K inhibitor alpelisib. The CDK4/6 inhibitors are effective in different treatment situations both in combination with fulvestrant or aromatase inhibitors [29 ], [30 ], [31 ], [32 ], [33 ], [34 ], [35 ], [36 ], [37 ]. The 3 studies that recruited hormone-sensitive patients and tested a combination with an aromatase inhibitor
(MONALEESA-2, MONARCH-3 and PALOMA-2) all had as a requirement that the patients had no progression on (neo-) adjuvant therapy and that the conclusion of the (neo-) adjuvant therapy had to be at least 12 months previously [32 ], [34 ], [36 ]. Against this background, these studies showed a clear benefit for disease-free survival even when an aromatase inhibitor was again combined with a CDK4/6 inhibitor in the metastatic situation. Nevertheless, real-world data suggest that fulvestrant is often preferred as anti-hormonal therapy in the first palliative treatment line for recurrence [25 ]. In this connection, the PARSIFAL study published at ASCO 2020 is of particular importance [38 ]. This study compared treatment with palbociclib + fulvestrant in this described hormone-sensitive situation (similar to MONALEESA-2,
MONARCH-3, PALOMA-2) with treatment with palbociclib-letrozole. The primary study aim was progression-free survival. 486 patients in total were included and randomised 1 : 1. A statistically significant difference was not found between the two arms with regard to progression-free survival (hazard ratio = 1.13 [95% CI: 0.89 – 1.45]). There was also no difference with regard to overall survival, where the hazard ratio was 1.0 (95% CI: 0.68 – 1.48) [38 ]. When interpreting these data, it must be borne in mind that this study initially investigated fulvestrant for superiority, which was not demonstrated. This study then looked at non-inferiority, which likewise was not confirmed. A treatment decision should always therefore depend on the overall view of the disease and the planned therapy sequences.
Alpelisib after CDK4/6 inhibitor treatment
In the Solar-1 study, which showed an improvement in PFS by addition of alpelisib in patients with a PIK3CA mutation of the tumour, only a few patients had received previous treatment with a CDK4/6 inhibitor. Only 7 patients received treatment with alpelisib and fulvestrant and 10 patients received placebo and fulvestrant. The median progression-free survival was 5.5 months in patients treated with alpelisib and 1.8 months in those who had been treated with fulvestrant alone [28 ]. It is therefore clear that better evidence was needed for this group of patients. This evidence was delivered by the recently presented BYLIEVE study in which patients were included who were to be treated again with anti-hormonal therapy following treatment with a CDK4/6 inhibitor [39 ]. This non-randomised study consisted of several cohorts, of which only cohort A has been presented to date. Cohort A included 112 patients
who had previously received treatment with a CDK4/6 inhibitor and an aromatase inhibitor. In the BYLIEVE study all patients received the combination of alpelisib and fulvestrant. Naturally, a tumour mutation in the PIK3CA gene had to be found in all patients. The primary end point was the number of patients who had no breast cancer disease progression after 6 months. This number was 50.4% (95% CI: 41.2 – 59.6). The median progression-free survival (secondary end point) was 7.3 months (95% CI: 5.6 – 8.3) [39 ]. It should be noted critically, however, that the BYLIEVE study is not a prospective randomised study that compares the combined treatment consisting of alpelisib and anti-hormonal therapy with a different therapy. Overall, treatment with alpelisib and fulvestrant could represent the preferred treatment option after therapy with a CDK4/6 inhibitor based on the available data for patients with a PIK3CA mutation in the
tumour. With regard to the side effects, it must be noted that clinically significant hyperglycaemic episodes and skin rashes occur in approximately one third of the patients, often necessitating interdisciplinary management.
Biomarkers
Tumour mutational burden and pembrolizumab
The KEYNOTE-119 study is a study in which patients with advanced TNBC in later treatment lines were treated either with pembrolizumab alone or with chemotherapy [40 ]. Even though the primary study aim, namely, superiority of pembrolizumab compared with chemotherapy, was not reached, the study showed clearly that pembrolizumab was all the more effective the higher the CPS score was (PD-L1 positivity in immune cells and tumour) [40 ]. The median overall survival was 14.9 months in patients with a CPS ≥ 20 and 12.7 months in patients with a CPS ≥ 10, 10.7 and 9.9 months with CPS ≥ 1, when no limitations were made with regard to the CPS scores [40 ].
The tumour mutational burden (TMB) was also discussed in this context as a biomarker for the response to immunotherapy [41 ], [42 ], [43 ], [44 ]. The TMB status of 253 of the 601 patients in the KEYNOTE-119 study was determined. 26 patients (10.3%) had a TMB ≥ 10 mutations per megabase pair (mut/Mb). The response rate with pembrolizumab was numerically higher (14.3%) in patients with a high TMB than in patients with a low TMB (12.7%). The difference between the two treatment arms (pembrolizumab vs. chemotherapy) with regard to overall survival also appeared to be greater in patients with a high TMB (hazard ratio: 0.58; 95% CI: 0.21 – 1.57) than in patients with a low TMB (hazard ratio: 0.81; 95% CI: 0.61 – 1.07).
Whether and how TMB can be established as a biomarker in patients with breast cancer remains to be seen.
Outlook
Tucatinib and trastuzumab-deruxtecan are two substances that are entering clinical practice for HER2-positive advanced breast cancer. This could further markedly improve the treatment situation of patients with a HER2-positive tumour. In patients with TNBC, a therapy advantage was shown with pembrolizumab for a second checkpoint inhibitor (after atezolizumab) in first-line therapy. With regard to treatment with PARP inhibitors an attempt is being made to expand the treatment indication to other mutations or HRD signals. For HER2-negative HR-positive tumours, following CDK4/6 inhibitor therapy, a further targeted therapy (PI3K inhibition) has entered clinical practice with alpelisib. The trend of the last few years is thus confirmed, showing that drug development is progressing at high speed and hardly a year passes in which new substances are not presented for all subtypes of metastatic breast cancer with promising new data from clinical studies.
