Key words early breast cancer - therapy - prognosis - immune therapy - digital medicine
Introduction
Care for patients with breast cancer is complex and incorporates prevention, early
detection, treatment and follow-up. Significant progress has been made in each of
these areas in recent years [1 ], [2 ], [3 ], [4 ], [5 ], [6 ], [7 ], [8 ], [9 ].
Evidence has been accumulating that risk prediction for the disease can now differentiate
relatively well between validated, genuine risk genes and genes from analytical panels
that have no straightforward association with breast carcinoma. As a result, a major
study to clarify the findings has been published.
The speed with which new therapies are being introduced is accelerating significantly.
For example, the effectiveness of the antibody drug conjugate sacituzumab govitecan
was recently demonstrated in a randomised study for triple-negative breast cancer
(TNBC), and studies on the tyrosine kinase inhibitor tucatinib and the antibody drug
conjugate trastuzumab-deruxtecan are currently being conducted for HER2-positive breast
cancer. In patients with HER2−/HR+ breast cancer, new targeted combinations and also
new anti-oestrogenic agents are being tested following the introduction of CDK4/6
inhibitors and the PI3K inhibitor alpelisib.
Significant developments in healthcare have also been made in the field of digital
medicine, partly as a result of the COVID 19 pandemic. This review summarises and
reflects upon such recent developments, as presented in scientific publications and
at recent congresses such as the ESMO Congress 2020.
Neoadjuvant Therapy
Immune therapies in neoadjuvant therapy
The KEYNOTE-522 study, a large randomised neoadjuvant study, showed that supplementation
of therapy with pembrolizumab in triple-negative breast cancer resulted in an improvement
in the pathological complete remission rate (pCR) from 51.2% to 64.8% [10 ], [11 ], [12 ].
Similar results have now been reported with atezolizumab [13 ]. The IMpassion031 study included patients with early TNBC whose primary tumour was
at least 2 cm in size. Another requirement was that patientsʼ PD-L1 status had to
be determinable. However, it did not have to be positive. Patients were treated with
chemotherapy of 12 cycles of weekly nab-paclitaxel followed by dose-dense therapy
with biweekly doxorubicin (60 mg/m2 ) combined with cyclophosphamide (600 mg/m2 ). The patients also received atezolizumab at 840 mg or placebo every 2 weeks. After
surgery, therapy with atezolizumab was completed with 11 three-weekly cycles of 1200 mg
in the atezolizumab arm. A total of 333 patients were randomised and 307 patients
underwent surgery. The pCR rate in the placebo arm was 41.1%, while the pCR rate in
the atezolizumab arm was 57.6% (16.5% difference, 95% CI: 5.9 – 21.7%, p = 0.0044).
In the subgroup analysis based on
PD-L1 status (positive if at least 1% of immune cells exhibited staining), pCR
rates of 68.8% (atezolizumab arm) and 49.3% (placebo arm) were observed in the group
with positive PD-L1 status (difference 19.5%, 95% CI: 4.2 – 34.8%). In patients with
PD-L1-negative breast cancer, pPCR rates of 47.4% (atezolizumab arm) and 34.4% (placebo
arm) were achieved (difference 13.3%, 95% CI: − 0.9 – 27.5%) [13 ]. Survival parameters (event-free survival, disease-free survival and overall survival)
with a median observation period of about 20 months were also reported. On the basis
of very few events and wide confidence intervals, the atezolizumab arm showed a numerical
(statistically non-significant) advantage over the placebo arm. No new side-effect
profiles were reported. Thus, in the neoadjuvant setting in TNBC, two randomised trials,
one on pembrolizumab [11 ] and the other on atezolizumab [13 ], have now shown a significant improvement in pCR rate.
[Table 1 ] provides an overview of the neoadjuvant, randomised trials that are investigating
therapy with a PD1 or PD-L1 inhibitor.
Table 1 Overview of neoadjuvant studies (reported and as yet unreported) involving a PD1
or PD-L1 inhibitor.
Name of study
Population
Number of patients
Date of first publication
Which therapies are compared with each other?
pCR rates in the study arms
Survival data from both arms
Reference no.
* Statistically not significant in accordance with the pre-specified analysis plan
including interim analyses
GeparNuevo
Triple-negative breast cancer
174
1 August 2019
Durvalumab or placebo in addition to nab-paclitaxel followed by EC
47/88 (54.4%) in the durvalumab arm vs. (44.2%) in the placebo arm
As yet unpublished
[54 ]
KEYNOTE-522
Triple-negative breast cancer
602
27 Feb. 2020
Pembrolizumab vs. placebo plus paclitaxel and carboplatin
64.8% in the pembrolizumab arm vs. 51.2% in the placebo arm
Event-free survival with an HR in favour of the pembrolizumab arm (HR = 0.63; 95%
CI: 0.43 – 0.93*)
[11 ]
KEYNOTE-756
Oestrogen receptor positive, HER2−
Planned: 1140
26 May 2019
Pembrolizumab vs. placebo with neoadjuvant chemotherapy and adjuvant endocrine therapy
As yet unpublished
As yet unpublished
[55 ]
NeoTrip
Triple-negative breast cancer
280
12 Dec. 2019
Atezolizumab vs. placebo with carboplatin and nab-paclitaxel
43.5% with atezolizumab vs. 40.8% with chemotherapy alone
As yet unpublished
[56 ]
IMpassion031
Triple-negative breast cancer
455
20 Sep. 2020
Atezolizumab vs. placebo with chemotherapy
95/165 (58%) vs. 69/168 (41%)
Event-free survival (HR = 0.76, 95% CI: 0.40 – 1.44)
Disease-free survival (HR = 0.74, 0.32 – 1.70)
Overall survival (HR = 0.69, 0.25 – 1.87)
[13 ]
GeparDouze
Triple-negative breast cancer
Planned: 1520
As yet unpublished
Atezolizumab vs. placebo with neoadjuvant chemotherapy
As yet unpublished
As yet unpublished
[57 ]
CheckMate 7A8
Hormone receptor positive, HER2-negative, postmenopausal
Planned: 136
As yet unpublished
Nivolumab, abemaciclib, palbociclib and anastrozole
As yet unpublished
As yet unpublished
[58 ]
APTneo
HER2-positive
Planned: 650
As yet unpublished
Atezolizumab vs. placebo with trastuzumab, pertuzumab, carboplatin and paclitaxel,
or sequential therapy with anthracycline
As yet unpublished
As yet unpublished
[59 ]
Adjuvant Treatment of Patients with Breast Cancer
Adjuvant Treatment of Patients with Breast Cancer
The last time a novel anti-hormonal therapy was introduced in the adjuvant setting
was almost 20 years ago. At that time, findings of an absolute difference in disease-free
survival of 2% comparing 5 years of anastrozole to 5 years of tamoxifen [14 ] led in 2002 to authorisation in the USA. The hazard ratio in the initial analysis
with a median follow-up period of 33.3 months was 0.83 (95% CI: 0.71 – 0.96) [14 ]. The hazard ratio in the final analysis with a 10-year follow-up was 0.86 (95% CI:
0.76 – 0.97) for the first 5 years (total population) and 0.83 (95% CI: 0.72 – 0.95)
for hormone receptor positive tumours [15 ]. It should be noted that the ATAC study included some patients with unknown hormone
receptor status (approximately 8%) [14 ]. Beyond this first study on the use of an aromatase inhibitor in the adjuvant setting,
mention should likewise
be made of the large number of other studies in the adjuvant setting that have
also investigated letrozole and exemestane [16 ], [17 ]. Regardless of the successes achieved by the introduction of aromatase inhibitors
almost 20 years ago, the majority of deaths in breast cancer patients occur in the
HR+/HER2− group of patients, as these make up the largest percentage of all breast
cancers. Improving the treatment of these patients would have a major impact on the
overall population. Following the promising data on CDK4/6 inhibitors (CDK4/6i) in
the metastatic setting, trials have also been initiated for all three CDK4/6i in the
adjuvant setting [18 ], [19 ], [20 ], [21 ], [22 ]. The first two of these studies (PALLAS and MonarchE) have now been presented at
the
ESMO Congress 2020 [23 ], [24 ].
PALLAS
The PALLAS study included 5760 patients who had to fulfil the following criteria (selection):
HR+/HER2−
UICC stage II or III
Completion of all primary therapy options
Primary diagnosis not more than 12 months ago
Initiation of adjuvant endocrine therapy (ET) less than 6 months previously
Most patients had a UICC stage of IIB (33.4%) or III (48.7%). A very high percentage
had received chemotherapy (82.6%). The percentage of patients with a negative nodal
status was 13.0% [23 ].
Patients were treated with endocrine therapy (ET) ± palbociclib for 2 years. Afterwards,
the scheduled endocrine therapy is completed as a monotherapy.
The results of this study were negative. In an interim analysis, a “futility” analysis
was performed after the 351 events pre-specified to terminate the study. At a median
follow-up period of 23.7 months, the hazard ratio was 0.93 (95% CI: 0.76 – 1.15) with
invasive relapse-free 3-year survival rates (iDFS) of 88.2% (palbociclib + ET arm)
vs. 88.5% (endocrine monotherapy arm).
MonarchE
The MonarchE study investigating abemaciclib yielded a different result. Patients
were included in this study if they had more than 3 affected lymph nodes or 1 – 3
affected lymph nodes and a tumour size of at least 5 cm, a grade 3 tumour or a Ki-67
index of ≥ 20%.
Again, most patients in this study had a UICC stage of IIB (13.8%) or III (approx.
72%). Patients received standard-of-care adjuvant endocrine therapy with or without
abemaciclib for 2 years. At the time of the first interim analysis, invasive disease-free
survival was assessed after 323 events and a median follow-up time of 15.5 months.
The analysis revealed a significant advantage in favour of the combination therapy.
The hazard ratio was 0.72 (95% CI: 0.56 – 0.92), with 2-year invasive relapse-free
survival rates of 92.2% and 88.7%, respectively. [Fig. 1 ] presents the Kaplan-Meier curve for the results. [Fig. 2 ] shows a comparison of the Kaplan-Meier curves of the ATAC study [14 ] and the MonarchE study [25 ] at the time of initial publication.
Fig. 1 Remote metastases-free survival in the MonarchE study (data from [24 ]).
Fig. 2 Comparison of disease-free survival in the ATAC study with a 33.3-month follow-up
and invasive disease-free survival in the MonarchE study with a 15-month follow-up
(data from [24 ]).
PENELOPE-B study
The PENELOPE-B study [26 ] – initially reported to be negative via a press release – is another adjuvant CDK4/6i
study. This study included patients who did not achieve pCR after neoadjuvant chemotherapy
and had an unfavourable prognostic profile based on a clinical-pathologic stage –
estrogen/grade (CPS-EG) score [27 ]. PENELOPE-B included patients who had a CPS-EG score of 3 or higher or a score 2
and ypN+ disease. The PENELOPE-B study was the only placebo-controlled study, and
therapy with palbociclib lasted 1 year. The final analysis of the study has now been
published [28 ].
NATALEE study
One study still open for recruitment is the NATALEE study [19 ]. This study is evaluating ribociclib therapy over 3 years in a population of patients
at a lower and higher risk of relapse. The number of patients was recently increased
from approximately 4000 patients [29 ] to 5000 patients [19 ]. In the PENELOPE-B and MonarchE studies patients were administered a CDK4/6 inhibitor
for one year or two years respectively. The NATALEE study, in contrast, will provide
information on a longer duration of therapy of three years. Another study, the ADAPTcycle
study, is being conducted in Germany, comparing adjuvant chemotherapy with endocrine-based
therapy (2 years of ribociclib), while the ADAPTlate study is investigating the use
of abemaciclib 2 – 6 years after initial diagnosis.
[Table 2 ] compares the inclusion and exclusion criteria of the various adjuvant CDK4/6 inhibitor
studies.
Table 2 Comparison of inclusion and exclusion criteria of (post-)adjuvant therapy studies
with CDK4/6 inhibitors.
Criterion
PALLAS
MonarchE
NATALEE
PENELOPE-B
ALT = alanine aminotransferase, AST = aspartate aminotransferase, ECG = electrocardiography,
GFR = glomerular filtration rate, INR = International normalized ratio, QTcF = QT
interval corrected with the Fridericia formula, ULN = upper limit of normal
Age
≥ 18 years
≥ 18 years
≥ 18 years
≥ 18 years
Tumour stage
AJCC stage II:
T0/T1 N1
T2 N0
T2 N1
T3 N0
or AJCC stage III
T1–T4 and N1 with
or
or
T3 N0 or T2 N0 with
G3 or
G2 and Ki-67 ≥ 20% or oncotype DX ≥ 26, PAM50 high risk, MammaPrint high risk, EndoPredict
high risk
≥ ypT1 or
≥ ypN1
After at least 16 weeks of neoadjuvant chemotherapy
Tumour biology
HR+/HER2−
HR+/HER2−
HR+/HER2−
HR+/HER2− and CPS-EG score of ≥ 3 or CPS-EG of 2 with N+
Study schedule
Randomisation within 12 months of initial diagnosis and a maximum of 6 months after
the commencement of endocrine therapy
Randomisation within 16 months of surgery and at least 21 days after the last chemotherapy
and at least 14 days after the final radiotherapy
Randomisation within 18 months of initial diagnosis and no more than 6 months after
the commencement of endocrine therapy or “high risk” on the basis of a gene expression
test (Onkotype DX, Prosigna, MammaPrint or EndoPredict)
Randomisation within 16 weeks of surgery or maximum 10 weeks after completion of radiotherapy
ECOG
≤ 1
≤ 1
≤ 1
≤ 1
ECG
QTcF < 480 ms
No limit specified
QTcF < 450 ms
QTcF < 480 ms
Treatment of Patients with Metastatic Breast Cancer
Treatment of Patients with Metastatic Breast Cancer
Immunotherapies IMpassion131, IMpassion130, KEYNOTE-355
The PD-L1 inhibitor atezolizumab has now been authorised for patients with advanced
triple-negative breast cancer following the findings of the IMpassion130 trial, which
revealed an improvement in progression-free survival and overall survival in patients
with immune cell (IC) PD-L1-positive tumours [30 ], [31 ]. Similarly, the KEYNOTE-355 study showed that the addition of pembrolizumab to chemotherapy
significantly improved progression-free survival. In the USA, pembrolizumab has already
been authorised for this indication.
The IMpassion130 study selected nab-paclitaxel as combination therapy for atezolizumab.
Various chemotherapies (nab-paclitaxel, paclitaxel, or gemcitabine and carboplatin)
were permitted in the KEYNOTE-355 study. Based on a subgroup analysis of combination
partners, the KEYNOTE-355 study found no differences between the chemotherapy options
with which pembrolizumab was combined.
The IMpassion131 study [32 ] investigated a study population similar to that of IMpassion130, but whose subjects
were randomised to paclitaxel + atezolizumab or to paclitaxel monotherapy (+ placebo)
at a 2 : 1 ratio [32 ]. A total of 651 patients were recruited, of which 292 had tested positive for PD-L1
(immune cells positive in ≥ 1%). Neither in the PD-L1-positive population nor in the
overall population was a benefit observed for progression-free survival or overall
survival [32 ].
It has been suggested that the lack of effect of atezolizumab in combination with
paclitaxel may have been due to cortisone, which was co-administered during therapy
with soluble paclitaxel. Ultimately, however, such an explanation must be regarded
as speculative. It is known that different chemotherapies have different effects on
the immune system [33 ] and also interact differently with PD1/PD-L1 inhibitors [34 ].
Convincing data on sacituzumab govitecan in TNBC patients (ASCENT study)
There have already been significant therapeutic successes in the field of antibody
drug conjugates (ADC) in HER2-positive patients [35 ], [36 ], [37 ]. An epithelial glycoprotein (Trop-2) that is expressed by breast cancer cells and
is associated with a poorer prognosis has now been identified in patients with triple-negative
breast cancer, as well as potentially in the future in other subtypes [38 ], [39 ]. The ADC sacituzumab govitecan acts on this target, with SN-38, a cytostatic agent
similar to irinotecan but highly potent, as a payload. In the USA, sacituzumab govitecan
has already been authorised on the basis of positive data from the early therapeutic
study [40 ].
ASCENT is a randomised phase III study in which TNBC patients with ≥ 2 prior chemotherapies
are treated with either sacituzumab govitecan or a chemotherapeutic treatment of physician
choice [60 ]. A total of 529 patients were included. The study was terminated early due to a
significant difference between the treatment arms.
All included patients had previously received a taxane, approx. 7 – 8% had received
a PARP inhibitor and approx. 26 – 29%, a PD1/PD-L1 inhibitor.
A clear difference was observed between the randomised arms of the study. Patients
receiving chemotherapy of physician choice progressed at a median of 1.7 months (95%
CI: 1.5 – 2.6), while patients receiving treatment with sacituzumab govitecan did
not progress until 5.6 months (95% CI: 4.3 – 6.3). The corresponding hazard ratio
was 0.41 (95% CI: 0.32 – 0.52, p < 0.0001). A clear difference in overall survival
was also observed. The median time to death for patients in the chemotherapy arm was
6.7 months (95% CI: 5.8 – 7.7), while the median time to death for patients in the
sacituzumab-govitecan arm was 12.1 months (95% CI 10.7 – 14.0). The hazard ratio was
0.48 (95% CI: 0.38 – 0.59, p < 0.0001). [Fig. 3 ] presents the Kaplan Meier curves.
Fig. 3 Progression-free survival (a ) and overall survival (b ) for randomisation arms in the ASCENT study (SG = sacituzumab govitecan; TPC = treatment
of physicianʼs choice. Data from [60 ], survival rates were added).
The most common side effects (all grades) were neutropoenia (63%), anaemia (34%),
vomiting (29%), diarrhoea (59%) and fatigue (45%). However, these led in only 4.7%
of patients to premature discontinuation of therapy.
Overall survival analysis of the SOLAR-1 study
The PI3K inhibitor alpelisib was recently approved after it had been shown to improve
median progression-free survival from 5.7 months to 11 months in patients with HER2−/HR+
metastatic breast cancer and with a somatic PIK3CA mutation who had previously received endocrine therapy (hazard ratio: 0.65, 95% CI:
0.50 – 0.85) [41 ]. Data on overall survival have now been reported revealing 181 patients died (out
of a total of 341 patients) [42 ].
Even if the addition of alpelisib to fulvestrant improved overall survival from 31.4
months (95% CI: 26.8 – 41.3) to 39.3 months (95% CI: 34.1 – 44.9), this difference
was not statistically significant (HR = 0.86, 95% CI: 0.64 – 1.15, p = 0.15). A subgroup
analysis suggested that a significant part of the effect was attributable to patients
with a lung or liver metastasis. In this subgroup of 190 patients, the difference
in median overall survival was almost 15 months (37.2 vs. 22.8 months, HR = 0.68,
95% CI: 0.46 – 1.00) [42 ].
As testing is one of the prerequisites for alpelisib therapy, testing and methodology
are increasingly a focus of interest. Mutations can be analysed from DNA extracted
from paraffin-embedded tumours as well as from circulating DNA (ctDNA). According
to its protocol [41 ], the SOLAR-1 study tested for the following mutations: C420R, E542K, E545A, E545D
(only 1635G>T), E545G, E545K, Q546E, Q546R, H1047L, H1047R and H1047Y. Alpelisib is
likely to be effective against tumours with a number of other different mutations,
but prospective data on this should be awaited. It should be noted, however, that
mutations other than those mentioned are very rare, as the vast majority of mutations
in PIK3CA are restricted to three so-called hotspots.
Focus on quality of life
Quality of life is a particular focus of treatment for patients with advanced cancer.
If possible, therapy should have a positive effect on quality of life. At a minimum,
the patientʼs quality of life should be maintained and not worsened. This can be achieved
by successful symptomatic treatment or by delaying progression. It has been established
that progression is associated with a deterioration in quality of life [43 ]. Treatment with novel substances is often associated with an intensification of
therapy or with the introduction of an additional concomitant drug. Critical assessment
of quality of life is therefore important. Several analyses on quality of life have
been published on CDK4/6 inhibitors that are widely used in early lines of therapy
in advanced breast carcinoma [44 ]. Most such analyses revealed combination therapy and endocrine monotherapy were
associated with similar quality-of-life scores
[45 ], [46 ], [47 ], while one study revealed a benefit for combination therapy [48 ]. Recently, a pooled analysis of all ribociclib studies was performed [49 ]. Reviewing all studies, this analysis showed that combination therapy can significantly
delay deterioration in quality of life. In subgroup analyses, which generate hypotheses,
the effect was greatest in patients between 45 and 60 years of age and in patients
with visceral metastases [49 ].
For patients with HER2-positive breast cancer who have already been treated with all
standard options, further treatment options with tucatinib and trastuzumab deruxtecan
will most likely be available in the future (after a pending authorisation in Europe)
[37 ], [50 ]. Quality-of-life data have recently been published for the randomised HER2CLIMB
study [51 ]. In this analysis, in which almost half of all patients had brain metastases, the
addition of tucatinib to trastuzumab and chemotherapy did not result in any difference
in quality of life [51 ] despite a more unfavourable side-effect profile [50 ]. It is worth noting that significant benefits in progression-free survival and overall
survival had already been reported for tucatinib when comparing the randomisation
arms [50 ].
Digital Medicine
Catalysed in part by the COVID-19 pandemic, the prospects for digital medicine have
once again undergone a transformation. The pandemic has highlighted the obvious advantages
of collecting data digitally and providing care to patients, without them having to
physically attend medical appointments in a hospital or in private practice. However,
it was clear even before the pandemic that shifting appointments to monitor therapies
and illnesses to the patientsʼ own homes had the potential to improve the quality
of life of patients and reduce the burden on the healthcare system [52 ]. In addition, modern smart sensory systems are enabling patients to collect medically
relevant information [53 ]. The ability to record ECG data, conduct blood analyses, monitor activity and sleep
patterns as well as other information via smart phones and smart watches and to use
this information for patient care is opening up new
perspectives in both care and research.
Collecting information about patientsʼ conditions allows caregivers to communicate
with patients about such information and to analyse the data to gain new insights
into patient groups. Data from private practice and from hospitals can be merged with
other healthcare-related data. Such datasets can be managed and analysed using machine
learning to conduct outcome-oriented research. The findings could enable compliance
with guidelines and improve the information provided to patients, patient care, home-based
care and medical research in equal measure. [Fig. 4 ] illustrates a possible network of this type.
Fig. 4 Perspectives for digital medicine.
Outlook
Some of the studies presented in this review are relevant to clinical practice; they
either describe therapies that for the first time are clearly effective in disease
settings where no effective therapies previously existed (sacituzumab govitecan in
patients with pretreated, advanced TNBC), or they answer specific issues of particular
interest to clinicians. Thus, the issue as to whether atezolizumab can also be successfully
combined with paclitaxel – disproved by the IMpassion131 study – has been resolved
for the time being. Atezolizumab should still be combined with nab-paclitaxel. The
IMpassion031 study has now resulted in data from a second, larger study in the neoadjuvant
therapeutic setting involving checkpoint blockade. It remains to be seen how rapidly
these therapies can be granted authorisation. Such authorisation would require demonstration
of a clear benefit with regard to the increased rate of side effects described above.
Finally, new methods of care, such as digital medicine, offer the prospect that healthcare
can increasingly be shifted into the home. It is hoped that scientific studies will
be conducted to establish the extent to which this will improve patientsʼ quality
of life. To date, the personal contact between doctors and patients has been the most
important factor in establishing good therapy compliance and trust in the necessary
therapeutic measures.
