Presented by: Donatienne Taylor, MD (CHU UCL Namur, Namur, Belgium)
From a theoretical point of view, treatment escalation strategies provide additional treatment to patients who require it, while de-escalation of therapy enables patients to forego unnecessary treatment, avoiding unnecessary treatmentrelated adverse events (TRAEs). The identification of patients who might need escalation or de-escalation requires biomarkers that are not only prognostic, but also predictive with a proven clinical utility. For post-menopausal endocrine receptor-positive (ER+)/ human epidermal growth factornegative (HER-) breast cancer (BC), gene expression profiling (GEP) has a proven its clinical utility in this regard.
Using the 70-gene MammaPrint platform, the phase III MINDACT trial demonstrated a comparable rate of distant metastasis free survival (DMFS) when adjuvant chemotherapy (ACT) was omitted in clinical risk high/genomic risk low HR+/HER2- BC patients (ACT 8-year DMFS: 90.2%. No ACT 8-year DMFS: 90.0%).21 Similarly, the phase III Rx-Ponder trial found that postmenopausal HR+/HER2- BC patients, who have 1–3 positive nodes and a 21-gene recurrence score (RS) of ≤ 25 could safely avoid adjuvant chemotherapy, finding significantly comparable invasive disease-free survival (IDFS) rates between patients who received chemotherapy + endocrine therapy (CET) or endocrine therapy (ET) alone (ET 5-year IDFS: 91.9%. CET 5-year IDFS: 91.6%. HR[95%CI]: 0.97[0.78–1.22], p= 0.82).22 Similar results were also found for node-negative disease patients in the TAILORx trial.23 Unfortunately, the use of GEP as a decision tool to de-escalate therapy does not seem to be feasible in pre-menopausal women, with the MINDACT trial finding an absolute 5% difference in DMFS with adjuvant chemotherapy was omitted (ACT 8-year DMFS: 93.6%. No ACT 8-year DMFS: 88.6%).21
Several other biomarkers have also been studies in patients with breast cancer, with the phase III Poetic trial demonstrating that a drop Ki67 below 10% after a short course of ET is prognostic in post-menopausal women.24 Establishing the clinical utility of this finding, the phase II/III ADAPT trial assessed Ki67 and RS in HR+/HER2- BC patients before receiving a short course of ET. In this trial, patients who had an RS of >25 and a Ki67 >10% post-ET were subsequently treated with chemotherapy. In contrast, patients who had an RS of 12–25 and a Ki67 <10% post-ET, and patients who had a pre-ET RS <12 were only treated with continued ET. With comparable 5-year IDFS rates for RS 12–25 and Ki67 10% post-ET compared to RS <12 patients, this trial showed that these patients can safely forego chemotherapy.25 However, dynamic Ki67 is unlikely to be more useful than GEP, mainly due to its low reproducibility. The ADAPT trial also did not enrol premenopausal women and so the applicability of dynamic Ki67 to these patients cannot be determined. Several neoadjuvant treatment strategies for ER+/HER- BC are currently under investigation.
Both the KEYNOTE 756 and B-IMMUNE trials are assessing the efficacy and safety of chemotherapy in combination with immune checkpoint inhibition in this setting.26,27 Also in the neoadjuvant setting, anti-ROS strategies are being investigated for lobular BC patients in the phase II ROSALINE trial.28 The efficacy of cyclin-dependent kinase 4/6 inhibitors (CDK4/6) in combination with aromatase inhibitors is being assessed in the FELINE and NeoPAL trials.29,30 In this respect, it is important to note that positive results have been reported with CDK4/6 inhibitors in the adjuvant treatment of ER+/HER- BC patients in the MONARCH-E and NATALEE trials.31,32 Representing a de-escalation approach, the APPALACHES trial is investigating whether CDK4/6 inhibition could replace chemotherapy in this setting.33 The breast cancer index (BCI) also shows promise as a predictive and prognostic tool, and may be used to determine which patients require escalation and de-escalation treatments.
Also in patients with HER2+ BC, escalation and de-escalation strategies have been explored. For example, node-positive patients were shown to benefit from the addition of pertuzumab to trastuzumab and chemotherapy.34 Furthermore, in case this treatment does not result in a pCR for the patient, the phase III KATHERINE trial demonstrated a clinical benefit from the use of trastuzumab emtansine (TDM1) (HR[95%-CI]: 0.50[0.39–0.64], p< 0.001).35 Neratinib may be another option for these patients also.36 For HER2+ BC patients with small tumours < 2cm (T1N0), twelve weeks of adjuvant paclitaxel followed by nine months of trastuzumab induces a low risk of early recurrence.37 Finally, T-DM1 has also been assessed in the early HER2+ BC setting, but was found to be inferior to the SoC-control of pertuzumab, trastuzumab and chemotherapy.38
Data also shows that anthracyclines can safely be omitted in certain stage II/III HER2+ BC patients. In fact, the phase III TRAIN-2 study found comparable early and late outcomes between patients who were given anthracycline-based or paclitaxel-based neoadjuvant regimens.39 Anthracyclines are associated with cardiac toxicities and a risk of leukaemia. As a result, omitting this drug class from treatment strategies will reduce patient toxicity exposure. Further evidence comes from the phase III BCIRG 006 trial, which found a non-anthracycline chemotherapy regimen to yield superior outcomes.40 Importantly, however, the TRAIN-2 study only included 34 patients with N2 or N3 disease, and anthracyclines may still prove efficacious in these more advanced settings. Furthermore, despite these therapeutic advancements, some HER2+ BC patients will still have an unmet clinical need. For example, T-DM1 has no effect on brain metastasis recurrence, and the KATHERINE trial reported a 3-year DFS of 83%, meaning that a substantial proportion of patients still has disease progression after a treatment with T-DM1. It is also possible for a patient with a pCR to relapse.
Looking forward, several trials are investigating therapeutic combinations that may be relevant for patients requiring further escalation. For example, the phase III IMpassion 050 trial is currently evaluating the role of immunotherapy when given in combination with neoadjuvant chemotherapy.41 Similarly, the phase III APTneo trial is investigating immunotherapy in combination with the HER2+ SoC of trastuzumab, pertuzumab and chemotherapy.42 The antibody-drug conjugate trastuzumab deruxtecan is also being compared to T-DM1 in this setting in the phase III DESTINY-BREAST-05 trial.43 Furthermore, the efficacy of T-DM1 in combination with tucatinib in preventing relapse is under evaluation in the phase III COMPASS HER2 RD trial.44 De-escalation treatment regimens are currently under investigation in this setting as well, with the phase II TOUCH trial investigating if chemotherapy can be safely substituted by a CDK4/6 inhibitor in post-menopausal women with ER+/HER2+ BC.45 With a similar protocol, the DECRESCENDO trial plans to open enrolment later in 2021.46
Triple-negative BC (TNBC) also has a clear unmet clinical need, with 50% of patients who do not achieve a pCR after neoadjuvant therapy relapsing within three years. For these patients, the addition of adjuvant capecitabine to a regimen that includes standard, neoadjuvant chemotherapy was determined to be safe and effective in the phase III CREATE-X trial.47 Maintenance capecitabine after standard adjuvant chemotherapy was also found to significantly improve DFS in the phase III SYSUCC-001 trial.48 The phase II/III GeparSixto trial concluded that the addition of carboplatin to taxane- and anthracycline-based chemotherapy improves pCR.49 However, the impact of platinum-based chemotherapy regimens on survival outcomes is unclear. Additionally, the phase III SASCIA trial is currently evaluating sacituzumab govitecan in comparison to a SoC chemotherapy regimen.50 Deescalation strategies may become applicable to this subgroup in the future, after a trial reported that tumour-infiltrating lymphocyte (TILs) concentration was prognostic in early TNBC and that patients could forego chemotherapy.51
Immunotherapy has also been investigated in TNBC. The addition of immune checkpoint inhibitors to neoadjuvant chemotherapy increases pCR rates by approximately 14–17%, as reported in the phase III KEYNOTE 522 and IMpassion 031 trials.52,53 However, currently there is no supportive survival data regarding these immunotherapy regimens in this setting. Finally, for TNBC patients with germline BRCA mutations, PARP inhibitors are under current investigation in this setting in the phase III OlympiA trial.54
BELG J MED ONCOL 2021;15(3):134-44