Key words early breast cancer - adjuvant therapy - neoadjuvant therapy - T-DM1 - Katherine -
prevention
Introduction
The prognosis of primary early breast cancer has continued to improve in the past
few decades. This is seen in the improvement in the 5-year survival as well as in
the increase in the rates of pathological complete remission (pCR) within the scope
of neoadjuvant therapy concepts [1 ], [2 ]. This suggests that not only optimisation of local therapy or early detection [3 ], [4 ] improved prognosis but also systemic therapy. The introduction of new substances
and therapeutic regimens was able to improve therapy in the (neo)adjuvant situation
little by little [5 ], [6 ], [7 ]. The choice of patient population appears to play an ever more important role here.
For example, some years ago, so-called post-neoadjuvant studies were initiated which
continued to treat patients who had not reached complete remission following neoadjuvant
therapy. This type of study in particular appears to play an important role nonetheless,
because they investigate a specific resistance population. This therapy concept but
also aspects of prevention, surgical treatment, radiation therapy and other treatment
strategies are discussed in this review.
Prevention and Risk Factors
Prevention and Risk Factors
Nearly 25 years after the discovery of BRCA1 and BRCA2 , the techniques for genotyping have developed considerably further while the costs
have decreased. Nowadays, when testing for risk genes in the genome, it is no longer
only BRCA1 and BRCA2 which are genotyped, but rather a number of other genes which also influence the
breast cancer risk. These are generally other genes which, in the case of a mutation,
are also associated either with a disease risk similarly high as that of BRCA1 or BRCA2 or those genes which lead to a moderate disease risk [8 ], [9 ], [10 ].
Various works have reported on the mutation frequencies or risks in comparison to
healthy control persons [11 ], [12 ], [13 ], [14 ], [15 ], [16 ], [17 ]. One of the genes which was discussed as being classified either in the high-risk
group (similar to BRCA1 and BRCA2 ) or in the group with a medium disease risk is PALB2
[11 ], [12 ]. Initial studies had estimated the lifetime risk between 35 and 55% [11 ], [18 ]. Another large study which analysed the extensive panel gene analyses on approx.
20 000 cases of breast cancer and 20 000 health control persons has now been published
[19 ]. This study describes the relative risks for BRCA1 and BRCA2 with values of 7.9 and 6.7 and shows a relative risk of 4.8 for PALB2 . Other genes identified with statistical significance were CHEK2 and ATM with relative risks of 2.5 and 1.7. If the cases were limited to the triple-negative
patients, odds ratios of approx. 40 for BRCA1 , approx. 14 for PALB2 and approx. 9 for BRCA2
[19 ] were seen. In clinical practice, lifetime risks are more helpful than relative risks.
The corresponding lifetime risks were calculated at 50 – 55% for BRCA1 and BRCA2. PALB2 followed at slightly below 35%. CHEK2 and ATM were below this figure, at 25 and 15% [19 ]. The lifetime risks appear to be high enough in order to discuss individual risk-reducing
measures, however they are not the only genetic factors known to increase the risk
of breast cancer. The high-grade and medium-grade penetrant risk genes explain about
20% of the familial risk for breast cancer, while low penetrant but frequent genetic
variants in over 170 loci explain a further 16% of the familial breast cancer risk
[8 ], [20 ], [21 ], [22 ], [23 ], [24 ], [25 ], [26 ], [27 ]. In order to also possibly harness these low penetrant risk variants for an individual
risk determination, a risk score with 77 gene loci has already been previously developed
[28 ]. This has now been supplemented with additional risk genes and redeveloped with
313 gene loci. Women in the highest percentile had a lifetime risk of approx. 33%
([Fig. 1 ]) which can by all means be relevant for individual counselling [29 ]. For women around 60 years of age, a 10-year disease risk of more than 10% can be
calculated [29 ]. As in the case of all risk calculations, the identification of women with a risk
for breast cancer with a poor prognosis is important. It was shown here that the polygenetic
risk score in particular predicted the risk for hormone-receptor-positive carcinomas.
While the lifetime risk for hormone-receptor-positive carcinomas could be calculated
at over 30%, the corresponding lifetime risk for hormone-receptor-negative carcinomas
was approx. 4% [29 ]. The genes which account for a subtype-specific risk are known in part [15 ], [22 ], [30 ], [31 ], [32 ], [33 ], [34 ], [35 ], [36 ], [37 ], [38 ], [39 ] and are of particular interest for the development of individualised preventive
measures.
Fig. 1 Cumulative lifetime risk for healthy women, as a function of a polygenic risk score
with 313 gene loci (from [29 ]). The percentiles of the risk score and the lifetime risk based on them are shown.
The additional combination with other risk factors could entail a further improvement
in the risk prediction, since it is known that non-genetic risk factors have subtype-specific
effects on the risk [40 ] and the polygenetic risk score either interacts with other risk factors or non-genetic
risk factors improve the risk prediction in addition to the risk score [41 ], [42 ], [43 ].
Surgical Treatment
This year, a panel of experts published a needs assessment which identified the fields
in which a special focus should be scientifically placed in the area of breast surgery
in the near future. The important objectives of the further development are shown
in [Table 1 ]: A roadmap for research needs in breast surgery at the current time [44 ]. For some of the issues raised in the report, there were interesting findings this
year on which future research approaches can be built.
Table 1 Important objectives on the further development of breast surgery (acc. to [44 ]).
Diagnosis and assessment
Neoadjuvant therapy
Surgical management
Special groups
Addressing overdiagnosis and overtreatment, particularly in the context of screening
Understanding the biology, importance, and long-term outcomes of intermediate (B3)
lesions and ductal carcinoma in situ
Understand the biological significance and optimal therapeutic strategies for additional
foci of previously subclinical disease detected with advanced imaging techniques
Finding the best model for symptomatic assessment that allows rapid, patient-centred,
and cost-effective evaluation, while maintaining diagnostic accuracy
Establishing how and when to stage for distant disease and how best to monitor for
relapse
Understanding which patients will benefit most from treatment with neoadjuvant chemotherapy
or endocrine therapy, how this benefit differs by biological subtypes, and what biomarkers
can best guide decision making (eg, through window-of-opportunity studies)
Identifying the optimal treatment choice and sequencing
Understanding the long-term outcomes of neoadjuvant endocrine therapy
Identifying the optimal modalities (radiological and biomarker) for monitoring treatment
response
Optimising rates of breast conserving surgery after neoadjuvant treatment
Identifying whether surgery can be safely omitted in patients who achieve a pathological
complete response (pCR), and what imaging or biopsy methods can reliably predict pCR
Understanding the optimal management of the axilla in patients undergoing neoadjuvant
therapy, particularly those who convert from node-positive to node-negative disease
during treatment and the role of sentinel lymph node biopsy after treatment
Developing strategies to reduce the rates of re-excision for patients undergoing breast
conserving surgery through developments in localisation techniques and intraoperative
margin assessment methods
Understanding the role of alternatives to surgical excision. Assessment of clinical
and cost-effectiveness of oncoplastic and reconstructive surgery using standardised
measures and how this is affected by patient factors and adjuvant treatment, particularly
adjuvant radiotherapy to the chest wall
Robust evaluation of novel procedures and techniques using appropriate methodology,
such as the IDEAL framework, and commitment of the surgical community to the concept
of no innovation without evaluation
Management of the axilla in patients with a positive sentinel lymph node biopsy
Defining and standardising indications for risk-reducing surgery and understanding
the long-term outcomes of bilateral mastectomy for those who are considered at high
risk
Defining and standardising indications for contralateral mastectomy in those with
previous cancer to optimise benefits and minimise harm
Increased understanding of breast cancer and differing patient needs in specific groups,
including patients younger than 40 years, older patients, men, and patients with pregnancy-associated
breast cancer
Further study of survivorship, including optimisation of follow-up, secondary prevention,
and the role of surgery for treatment-related morbidity, such as lymphoedema
Understanding the role of surgery for metastatic disease
Between 2006 and 2016, the Young Womenʼs Breast Cancer Study [45 ] which was conducted in the USA, included a total of 1302 women under age 40 with
invasive breast cancer, 317 of whom received neoadjuvant therapy. Pretherapeutically,
only 85 patients (27%) were judged to be candidates for breast-conserving therapy.
Posttherapeutically, this figure increased to 163 (51%). Only 80 of these patients
(49%) opted for breast-conserving therapy, 83 (51%) chose mastectomy. The two most
important reasons for a mastectomy were patient preference (46%) and/or a BRCA1/2 or TP53 mutation (37%). Of the 75 patients (24%) who achieved pCR, 48 (64%) received a mastectomy
and only 21 of them 21 (44%) for anatomical reasons (inflammatory carcinoma, extensive
intraductal components, etc.) [45 ]. These data show that, especially in young patients, the decision for or against
a mastectomy after neoadjuvant therapy is often made more for personal and risk-reduction
reasons than for strictly oncological reasons. Whether these results can be transferred
to other care structures, such as in Germany, has not yet been investigated to date.
In this context, reference should be made to the results of two other studies, each
with far more than 500 patients and which addressed with the long-term quality of
life following breast cancer surgery: the E5103 study [46 ] which had included all age groups, and another large multicentre study which assessed
quality of life (QoL) in patients under age 40 [47 ]. In both investigations, the authors found indications that the long-term quality
of life was negatively affected by the radical nature of the surgical approach. In
particular in the investigation which had included patients under age 40 (range 26 – 40,
mean age 37 years), the psychosocial and sexual well-being was significantly worse
in the group of patients who underwent mastectomy [46 ], [47 ]. It is known from other studies that dissatisfaction with the outcome following
a mastectomy without reconstruction persists for many years [48 ]. Such data must be continuously reviewed in light of modern and less traumatising
reconstruction techniques; however, they should also be mentioned within the context
of informed consent during preoperative counselling.
Several translational analyses of the SENTINA study [49 ] were presented this year on the question of management of the axilla following neoadjuvant
chemotherapy. In one investigation, the post-therapeutic involvement of axillary lymph
nodes in the case of affected sentinel lymph nodes prior to neoadjuvant therapy was
analysed. 71 out of 318 patients (22.3%) still had affected lymph nodes following
neoadjuvant therapy, whereby patients with a positive HER2 status and a negative axillary
status had the highest pCR rates of the breast [50 ]. In another analysis as well in which a normogram for the prediction of nodal conversion
was developed for patients with pretherapeutically affected lymph nodes, the greatest
predictive factor was the tumour biology [51 ]. These investigations make current concepts appear promising with regard to forgoing
axillary surgery in studies on patients with an aggressive tumour biology and pCR
in the breast in the case of post-therapeutic clinically unremarkable lymph nodes.
Radiation Therapy
Management in the case of positive lymph node involvement
The sentinel lymph node biopsy (SNB) is the standard in clinically unremarkable axillary
lymph nodes. However, what should be done if these lymph nodes are affected by tumour?
The ACOSOG0011 study showed that dispensing with a further axillary lymphadenectomy
(ALND) does not lead to an increased rate of recurrence, although 23% of patients
have other affected lymph nodes which remain in situ. The main critical points of
the study were the low statistical power (discontinuation due to low recruitment)
and the unclear irradiation fields at the axilla [52 ].
The main question of the AMAROS study (n = 1425) [53 ] was more clearly defined here: in the event of a positive SNB, should irradiation
(AxRT) or surgery (ALND) be performed? After 10 years of follow-up, a very low rate
of local recurrence in both arms was seen overall, although additional metastases
were found in the surgical arm in 32.8% of patients. The rate of axillary recurrence
was 1.82% in the AxRT arm and 0.93% in the ALND arm (HR 1.71; 95% CI: 0.67 – 4.39,
p = 0.365). In DFS as well, there was no difference (HR 1.19; 95% CI: 0.97 – 1.45).
However, the rate of lymphoedema requiring treatment was significantly higher in the
ALND arm. 82% of the patients received breast-conserving surgery and 17% underwent
mastectomy and thus the results for both collectives appear representatives with a
very low event rate, however. Conclusion: If axilla is clinically unremarkable and
despite affected sentinel lymph nodes, further surgery is not felt to be appropriate.
Whether extensive (AMAROS) or tangential (ACOSOG0011) irradiation should be performed
cannot be answered yet [53 ].
Partial breast irradiation
In radiation therapy as well, de-escalation is an important strategy for reducing
therapy-related morbidity and/or the duration of treatment. Partial breast irradiation
by means of interstitial brachytherapy, three-dimensional conformal external irradiation
or intraoperative irradiation (e.g. Intrabeam® ) could contribute to this. Within the scope of the randomised phase III study NSABP
B-39, which included a total of 4216 patients with primary breast cancer in stage
I – III, the non-inferiority of partial breast irradiation versus conventional whole-breast
irradiation was investigated [54 ]. All forms of partial breast irradiation were permitted. The ipsilateral rate of
recurrence was selected as the primary endpoint of the study. The mean follow-up was
10.2 years. The non-inferiority unfortunately could not be demonstrated, even though
the 10-year rate of recurrence in the case of partial breast irradiation was only
0.7% higher (4.6 vs. 3.9%). The recurrence-free interval in the case of partial breast
irradiation was in fact significantly shorter (recurrence-free 10-year interval 91.8
vs. 93.4%), however no difference was seen in the case of metastasis- and disease-free
survival or overall survival. The grade 3 – 5 rates of toxicity do not differ significantly.
Thus for the low-risk patients, partial breast irradiation may represent an option
due to the only slightly increased risk of recurrence versus whole-breast irradiation.
Irradiation of the lymphatic vessels
The indication for irradiation of the lymphatic drainage area (LDA) is based on the
current guidelines and therapeutic recommendations for the involvement of more than
three lymph nodes (LN), independent of the size of the tumour as well as high-risk
constellations (1 – 3 LNs affected, G2–3, ER/PR negative) [55 ], [56 ], [57 ]. A current meta-analysis which altogether included data from 13 500 patients from
14 studies, confirmed this approach [58 ]. While earlier studies from 1961 – 1978 showed a slightly improved breast cancer
mortality (− 0.5%) and had an increased overall mortality, the more recent studies
from 1989 and later demonstrated a significantly reduced breast cancer and overall
mortality (− 2.8% and − 2.9%). This can most likely be attributed to precision radiation
therapy which minimises the cardiac radiation exposure (below 8 Gy). In the subgroup
evaluation, patients with more than three affected LNs particularly benefited from
irradiation of the LDA. Thus the meta-analysis confirms the currently recommended
approach.
Therapy for Primary Triple-Negative Breast Cancer
Therapy for Primary Triple-Negative Breast Cancer
The treatment of triple-negative breast cancer (TNBC) in the adjuvant or neoadjuvant
situation is marked by the fact that chemotherapy demonstrates good efficacy in a
portion of the patients and this results in a considerable improvement in the prognosis.
Thus it was able to be shown in neoadjuvant studies that triple-negative patients
who achieve pCR have an excellent prognosis, similarly to HER2-positive patients [59 ], [60 ], [61 ], [62 ], [63 ], [64 ], [65 ], [66 ].
In the event of a lack of pCR following neoadjuvant chemotherapy, the CREATE-X study
examined an adjuvant therapy with capecitabine in HER2-negative patients [67 ]. Particularly in the triple-negative patients, this study, which was conducted in
Asia, showed an advantage for disease-free survival (DFS) with a hazard ratio of 0.59
(95% CI: 0.39 – 0.87) [68 ].
The CIBOMA/2004-01_GEICAM/2003-11 study was conducted in a different study setting
but with the same question regarding modified therapy [69 ]. This study, which was conducted in Spain and Latin America, admitted triple-negative
patients following adjuvant or neoadjuvant chemotherapy who received further treatment
after completing therapy with capecitabine or who did not receive any further therapy.
As expected, the toxicity in the experimental arm was higher. In addition, no improved,
recurrence-free survival could be observed (HR: 0.82 [95% CI: 0.63, 1.06], p = 0.136).
A difference could be detected only in a subgroup with non-basal TNBC carcinomas (EGFR
and CK5/6 negative) (p = 0.020, HR: 0.53 [95% CI: 0.31, 0.91]). However, since the
study was negative overall, it was also concluded in the subsequent discussion that,
outside of the conditions in the Create-X study, the use of capecitabine is not indicated
in patients with TNBC [69 ].
In some studies, the efficacy of gemcitabine, nab-paclitaxel and carboplatin in early
breast cancer has already been investigated [70 ], [71 ], [72 ], [73 ], [74 ], [75 ], [76 ], [77 ], [78 ], [79 ], [80 ], [81 ]. In summary, these studies showed that the addition of gemcitabine to standard therapy
did not lead to any improvement and the latter yielded a benefit in a comparison between
standard therapy and therapy containing platinum. In the ADAPT study, a higher pCR
rate (26 vs. 45%) in patients with therapy containing platinum could be found [82 ] in triple-negative patients following neoadjuvant chemotherapy in a comparison between
treatment with nab-paclitaxel and gemcitabine vs. nab-paclitaxel and carboplatin [82 ]. With regard to disease-free survival (DFS), the study did not find any difference
in the two treatment arms [83 ]. The question of possible predictive markers was posed in a recently presented analysis
[84 ]. While patients with pCR and higher PD1 expression had the best prognosis, no predictive
markers for the superiority of carboplatin in TNBC in neoadjuvant chemotherapy were
able to be identified. In patients with pCR after 12 weeks and a high baseline PD1
(mRNA), the postoperative continuation of chemotherapy with 4 cycles of epirubicin
and cyclophosphamide did not lead to a better prognosis. However, the decision regarding
continuation of the neoadjuvant therapy was not randomised. The authors evaluated
this as an indication for a possible future basis for de-escalation, even if the results
currently only generate hypotheses and cannot be assessed as the current basis for
decision-making [84 ].
Adjuvant Therapy of Primary, Hormone-Receptor-Positive, HER2-negative Breast Cancer
Adjuvant Therapy of Primary, Hormone-Receptor-Positive, HER2-negative Breast Cancer
There are primarily three questions associated to date with the therapy of hormone-receptor-positive,
HER2-negative breast cancer patient in the adjuvant situation: In which risk constellation
must chemotherapy be administered? What is the optimal anti-endocrine therapy? And
how long should this be given?
With regard to the question of chemotherapy, it is known that patients with a positive
hormone receptor status, particularly with low proliferation, do not respond well
to chemotherapy [60 ], [62 ], [85 ]. The question thus arises as to whether chemotherapy is of any use at all in such
a patient population. The TAILOR-X study recently showed that patients who had achieved
an intermediate score with regard to the risk of relapse in a multi-gen assay do not
benefit from adjuvant chemotherapy followed by antihormonal therapy in comparison
to antihormonal therapy alone [86 ]. Thus in this patient population, chemotherapy could be omitted. Newly presented
quality-of-life data from the TAILOR-X study highlight this therapeutic decision approach
[87 ] (further discussion in [88 ]).
In the question regarding the length of the adjuvant, antihormonal therapy with aromatase
inhibitors, therapy until the 10th year after diagnosis is recommended to date in
the guidelines and therapeutic recommendations more for patients with an increased
risk of relapse than for patients with a low risk of relapse. The numbers of cases
for such analyses were relatively small in the respective studies, however. The question
also arises as to whether expanded adjuvant therapy with an aromatase inhibitor after
tamoxifen brings as much benefit as after an aromatase inhibitor. These questions
were addressed by a meta-analysis of the Early Breast Cancer Trialists Collaborative
Group with more than 22 000 patients from 11 studies [89 ].
The very comprehensive analyses investigated, on the one hand, the effect of aromatase
inhibitors after 5 years of tamoxifen, after 5 years of aromatase inhibitors or after
5 – 10 years of a sequence of tamoxifen and aromatase inhibitors. In addition, subgroup
analyses were performed in the overall population for patients with 0, 1 – 3 and more
than 3 affected lymph nodes. The therapeutic effect was the greatest in the group
of patients who were pretreated only with tamoxifen and only marginal for patients
who had received five years of pretherapy with aromatase inhibitors. The relative
risks for all analyses are shown in [Table 2 ].
Table 2 Risk reductions of expanded antihormonal therapy with aromatase inhibitors (AI) after
tamoxifen (TAM), AI or TAM, followed by AI (according to [89 ]).
Prior therapy
n
Any relapse
Distant metastases
Breast cancer mortality
RR (95% CI)
p value
RR (95% CI)
p value
RR (95% CI)
p value
1 not reported
5 years TAM
7 483
0.67 (0.57 – 0.79)
< 0.00001
0.77 (0.63 – 0.93)
0.008
0.77 (0.59 – 1.00)
0.05
5 years AI
3 322
0.76 (0.61 – 0.95)
0.2
0.78 (0.59 – 1.04)
0.09
0.99 (0.68 – 1.44)
0.97
5 – 10 years Tam, then AI
11 387
0.82 (0.73 – 0.93)
0.002
0.92 (0.80 – 1.07)
0.29
0.93 (0.77 – 1.12)
0.45
All patients
22 192
0.76 (0.70 – 0.83)
< 0.00001
0.85 (0.77 – 0.95)
0.004
0.89 (0.77 – 1.02)
0.09
Pat. with N0
10 620
0.82 (0.71 – 0.95)
0.009
–1
–1
Pat. with 1 – 3 LN
6 919
0.74 (0.64 – 0.85)
0.00003
–1
–1
Pat. with > 3 LN
1 621
0.71 (0.56 – 0.89)
0.003
–1
–1
In the analysis of the relative risks for a recurrence as a function of the node status,
it was shown that the greatest effect could be seen in the population of patients
who had more affected lymph nodes at primary diagnosis ([Table 2 ]) [89 ]. It is also important to note that the risk of bone fractures due to the expanded
AI therapy was increased by 24% [89 ].
In a similar context, the AERAS study presented by Ohtani et al. is noteworthy: The
expanded therapy with anastrozole for a total of 10 years in 840 patients reduced
the DFS events by half in comparison to 843 patients whose endocrine therapy was ended
after 5 years (HR 0.548, p = 0.0004). No influence on overall survival was able to
be shown. At the same time, the fracture rate of 2.8% in the expanded therapy arm
was twice as high as in the control arm (1.1%) [90 ].
Another option for intensifying the adjuvant endocrine therapy is to combine the endocrine
therapy with substances which have already shown in a metastatic situation that they
can overcome endocrine resistance in at least some patients. After the introduction
of everolimus in the treatment of patients with metastatic breast cancer [91 ], [92 ], adjuvant studies were also subsequently started (e.g. NCT01674140, NCT01805271);
they are still awaiting publication. Another option is the combination with CDK4/6
inhibitors which have a more favourable adverse effect profile. In this regard, there
were recently meaningful results from the neoadjuvant therapy situation. Dowsett et
al. presented the results from the Pallet study: In this study, palbociclib was given
in addition to three months of neoadjuvant endocrine therapy with letrozole. It was
shown that the antiproliferative effect of the aromatase inhibitor is substantially
increased by palbociclib: The percentage of tumours which underwent a complete cell
cycle arrest in the form of a Ki-67 value < 2.7% during neoadjuvant therapy was able
to be increased through the addition of palbociclib from 58.5 to 90.4% [93 ].
With new, effective combination therapies, additional options are available which
increasingly improve the adjuvant therapy of the hormone-receptor-positive, HER2-negative
patient. At present, adjuvant therapy studies are being conducted for all CDK4/6 inhibitors
(Penelope, PALLAS, MonarchE and NataLEE).
With the further development of adjuvant antihormonal therapy, the question of compliance
arises, particularly in the case of an adverse effect profile known to be more unfavourable,
and this question has already been discussed in the adjuvant studies with an antiendocrine
monotherapy. Some studies have reported on adherence [94 ], [95 ], [96 ], [97 ], [98 ], which was between 60 and 90%. It will be of interest to see how this is influenced
by combination with a CDK4/6 inhibitor, particularly as it is known that adverse effects
are one of the main predictors for non-adherence.
Therapy of Primary HER2-positive Breast Cancer
Therapy of Primary HER2-positive Breast Cancer
Benefits of neoadjuvant therapy
Neoadjuvant systemic therapy permits in-vivo sensitivity testing in addition to a
reduction in surgical morbidity (more breast conservation, fewer axillary lymphadenectomies)
[99 ], [100 ]. Based on the effect of the neoadjuvant systemic therapy on the primary tumour,
its effect on the long-term prognosis can be estimated, possibly through the destruction
and monitoring of micrometastases [62 ], [101 ].
A recent meta-analysis once again highlighted the prognostic significance of reaching
pathological complete remission (pCR) following neoadjuvant chemotherapy [102 ]. After evaluating 52 studies (51.1% randomised; 6.1% single-arm; 42.8% retrospective)
with 27 895 patients and a median follow-up period of 4 years, it was confirmed that,
by achieving pCR, the risk of a breast cancer event decreases significantly by 69%
(HR 0.31; 95% CI 0.24 – 0.39) and the risk of dying decreases significantly by 78%
(HR 0.22; 95% CI 0.15 – 1.30). The absolute effect after 5 years on DFS and overall
survival (OS) was 21 and 19%, respectively ([Fig. 2 ] and [3 ]). With a short follow-up time, the absolute effect was the greatest in the case
of patients with triple-negative breast cancer, followed by patients with HER2-positive
and hormone-receptor-positive, HER2-negative breast cancer (Δ in the 5-year EFS 33
vs. 23 vs. 9%). According to the statistics, a Δ in the pCR rate of 20% transferred
in the studies into a reduction in the event risk by approx. 20% [102 ]. Additional postoperative chemotherapy after reaching pCR did not improve the prognosis.
Fig. 2 Event-free 5-year probability after neoadjuvant chemotherapy and with pathological
complete remission (blue) or without pCR (red).
Fig. 3 5-year overall probability after neoadjuvant chemotherapy and with pathological complete
remission (blue) or without pCR (red).
Improvement in prognosis through a switch to T-DM1 in the case of non-pCR
The phase III CREATE X study showed for the first time that, by adapting the postoperative
therapy to the pathological response to the neoadjuvant therapy, the risk of recurrence
and mortality can be significantly decreased [67 ]. While the CREATE X study included only patients with a HER2-negative breast cancer
who did not achieve pCR through neoadjuvant chemotherapy, the KATHERINE study tested
the same approach in patients with HER2-positive breast cancer [103 ], [104 ]. This study included 1486 patients with primary HER2-positive breast cancer who
had not achieved pCR following neoadjuvant standard therapy with at least one taxane
and trastuzumab for at least 9 weeks. The neoadjuvant therapy could include anthracyclines
and a dual anti-HER2 blockade. The patients were randomised postoperatively and received
either trastuzumab emtansine (T-DM1) 3.6 mg/kg or trastuzumab 6 mg/kg every 3 weeks
for 14 cycles, at the same time as locoregional and, in the case of hormone receptor
expression, endocrine standard therapy. Prospective stratification was performed according
to operability (primarily operable vs. inoperable), hormone receptor status (positive
vs. negative), type of neoadjuvant anti-HER2 therapy (trastuzumab vs. dual blockade
with trastuzumab and pertuzumab) and the nodal status following surgery (ypN0 vs.
ypN+). With a median follow-up period of 41 months, the switch to T-DM1 significantly
improved the primary endpoint, the invasive disease-free survival after 3 years (IDFS),
from 77.0 to 88.3% (Δ 11.3%; HR 0.50; 95% CI 0.39 – 0.64; p < 0.0001) ([Fig. 4 ]). The relative effect was the same in all stratified subgroups, particularly also
in the case of patients with a very small residual tumour (≤ ypT1b ypN0) and in those
who had neoadjuvantly received a dual anti-HER2 blockade. This also appears to be
important to mention for this reason, because the neoadjuvant therapy with trastuzumab
and pertuzumab, similar as in clinical studies, had also shown a higher pCR rate in
real-world analyses [105 ]. The metastasis-free survival (distant disease-free survival, DDFS) after 3 years
was also significantly improved from 83.0 to 89.7% (Δ 6.7%; HR 0.60; 95% CI 0.45 – 0.79).
This benefit was achieved at the expense of a clinically easily controlled increase
in thrombopenia (grade ≥ 3 Δ 5.7%), increased liver values (grade ≥ 3 Δ approx. 1%)
and polyneuropathy (grade ≥ 3 Δ 1.4%) [103 ]. Thus the switch to T-DM1 in the case of non-pCR following adequate neoadjuvant
systemic therapy in HER2-positive primary breast cancer represents a new therapeutic
standard.
Fig. 4 Invasive disease-free survival when comparing the two randomisation arms of the Katherine
study (modified according to [104 ]).
Duration of the trastuzumab treatment
In patients with HER2-positive, primary breast cancer who are indicated for treatment
with trastuzumab, the question repeatedly arises as to whether a one-year treatment
duration is absolutely necessary or whether shorter therapy can be considered [106 ]. In this context, the final survival data from the phase III PHARE study were recently
presented [107 ]. In this non-inferiority study, 3384 patients with HER2-positive, primary breast
cancer (57.7% hormone-receptor-positive; 44.6% nodal-positive; approx. 43% trastuzumab
therapy sequentially) who were still event-free after 6 months on trastuzumab randomly
received either trastuzumab for another 6 months or no further anti-HER2 therapy.
With a median follow-up time of 7.5 years, the DFS as well as the DDFS and OS after
only 6 months of trastuzumab therapy were not clearly equally good as after one year
of trastuzumab. The upper limit of the 95% CI of the HR was above the predefined maximum
value of 1.15. However, in the subgroup in which trastuzumab was already started during
chemotherapy, both therapy arms were equally effective. Nevertheless, the equivalence
of 6 vs. 12 months of trastuzumab treatment could not be demonstrated with sufficient
certainty overall and thus trastuzumab therapy for a year remains the standard.
Outlook
With the KATHERINE study, a large adjuvant study which can demonstrate a significant
reduction in the risk of relapse for the HER2-positive patient population treated
neoadjuvantly was presented. This is significant not only for this patient group but
also for patients with other tumour biologies. Patients with hormone-receptor-positive,
HER2-negative breast cancer were treated in a similar, post-neoadjuvant therapy concept
with palbociclib. The initial results are expected in mid-2019. The Olympia study
also included post-neoadjuvant patients with a BRCA mutation for therapy with olaparib.
Independent of the post-neoadjuvant situation, three large adjuvant studies with CDK4/6
inhibitors which also have a potential for significant therapeutic efficacy are currently
being conducted.
