Clinico-pathological predictors of radiologic complete response to first-line anti-HER2 therapy in metastatic breast cancer
Introduction
With more than 2 million new diagnoses per year, breast cancer (BC) represents the most common neoplasia diagnosed worldwide and the leading cause of cancer-related death1. In up to 10% of the patients, it occurs as a de novo metastatic disease, while it is estimated that about 30% of cases initially diagnosed as early stage, will later recur at distant sites2.
The expression of the ERBB2 proto-oncogene is observed in approximately 20% of metastatic breast cancers (MBC), defining a molecular subtype that has been conventionally characterized by worse survival outcomes2. However, the advent of anti-HER2 agents has dramatically altered the history of this disease, with the monoclonal antibody trastuzumab being the first targeted therapy approved in this setting in virtue of the consistent survival benefit observed when added to conventional chemotherapy (CT) compared to CT alone3.
In the phase 3 CLEOPATRA trial, the combination of docetaxel and the dual anti-HER2 blockade with trastuzumab-pertuzumab was found to improve survival outcomes in patients with HER2-positive MBC with respect to docetaxel and trastuzumab only, thus becoming the standard of care as first-line treatment for these patients4. As reported by the primary results of this trial, 19 patients (6%) allocated in the pertuzumab arm experienced a complete response (CR)4. Similar outcomes were reported in a retrospective study by Yeo et al., where 8% of patients receiving CT and trastuzumab-pertuzumab achieved a CR, which has been found to be associated with more favorable survival outcomes5,6.
To date, there is limited data regarding which factors could be predictive of a CR to anti-HER2 therapies and that might inform future de-escalation strategies in the maintenance setting.
Thereby, aim of this study was to describe a cohort of patients with HER2-positive MBC to identify clinico-pathological characteristics predictive of achieving a radiological CR (rCR) to a first-line anti-HER2 therapy and to assess the impact of rCR on overall survival (OS).
Methods
Study design and study population
This is an exploratory analysis of the GIM14 BIO-META study (NCT02284581), an observational retrospective/prospective study aiming to evaluate the number and duration of therapeutic regimens employed for the management of MBC regardless of the molecular subtype. The retrospective cohort includes patients diagnosed with MBC from January 2000 to the beginning of the study, while the prospective one includes patients treated from the study start-up in each center (April 2016 for the coordinating center) and collected data prospectively. Non-consecutive patients with MBC were excluded.
The present analysis focused on describing the clinico-pathological characteristics of patients with HER2-positive MBC treated with first-line anti-HER2 therapy from year 2000 to 2021. HER2-positive disease was defined as a HER2 Immunohistochemistry (IHC) score of 3+ or a HER2 IHC score of 2+ with gene amplification confirmed by in situ hybridization (ISH). HER2 status was assessed locally. Radiological evidence of metastatic involvement at the time of MBC diagnosis was performed per local practice, with imaging studies including full-body CT scans and bone scintigraphy (for patients with bone disease). Imaging restaging during the first and further lines of treatment was also performed according to local practice.
Patients were classified according to the best radiologic response obtained from first-line therapy based on anti-HER2 agents plus CT backbone, assessed locally through medical interpretation of radiological reports, and time-to-treatment-discontinuation (TTD). TTD was defined as the time from the beginning of treatment to its discontinuation for any cause, including disease progression, toxicity, or death. Radiologic complete response (rCR) was defined as a complete response with TTD > 3 months. A minimum 3-month treatment response was considered in our analysis since, in clinical practice, radiological re-evaluations are mostly performed after at least 3 months of treatment7.
Additionally, we investigated the variables associated with higher odds of achieving a sustained CR, which we defined as a CR with a TTD > 18 months.
Ethical approval
The GIM14 BIO-META study was approved by the ethical and local institutional review boards for the clinical trial sites, and it was carried out in accordance with Good Clinical Practice guidelines and the Helsinki Declaration.
The following independent ethical committees provided study approval: Comitato Etico Unico Regionale FVG (Aviano/Udine), Comitato Etico Regionale della Liguria (Genova), Comitato Etico Centrale IRCCS Lazio Sezione IFO-Bietti (Roma), Comitato Etico ASST Settelaghi (San Fermo della Battaglia), Comitato Etico Area Vasta Nord Ovest c/o A.O.U. Pisana (Pisa), Comitato Etico dell’IRCCS Istituto Nazionale Tumori Fondazione Giovanni Pascale (Napoli), Comitato Etico per le attività Biomediche Carlo Romano dell’Università degli Studi di Napoli Federico II (Napoli), Comitato Etico Internazionale dell’ASO S. Croce e Carle (Cuneo), Comitato Etico IRCCS di Candiolo (Candiolo), Comitato Etico Milano Area (Legnano), Comitato Etico Indipendente Azienda Ospedaliero-Universitaria Consorziale Policlinico di Bari (Bari), Comitato Etico dell’ IRCCS Fondazione “S. Maugeri” (Pavia), Comitato Etico dell’Università “Sapienza” (Roma), Comitato Etico della ASL 1 di Sassari (Sassari), Comitato Etico Interaziendale c/o P.O. Molinette (Torino), Comitato Etico A.O.R.N. “A. Cardarelli/Santobono-Pausilipon” (Napoli), Comitato Etico Indipendente della A.O.U. di Cagliari (Cagliari), Comitato Etico Lazio 2 (Sora/Frosinone), CE Lazio 1 (Roma), Comitato Etico Area Vasta Emilia Centro (Ferrara), Comitato Etico Interaziendale Milano Area A (Milano), Comitato Etico Lazio 2 (Roma), CET Interaziendale AOU Città della Salute e della Scienza di Torino (Torino), Comitato Etico dell’Università Cattolica del Sacro Cuore – Policlinico Universitario A. Gemelli (Roma), Comitato Etico Milano Area 2 (Milano), C.E.ROM. Comitato Etico della Romagna (Meldola), Comitato Etico dell’IRCCS dell’Istituto Clinico Humanitas (Rozzano), Comitato Etico dell’Area Vasta Emilia Nord (Parma), Comitato Etico Interaziendale dell’A.O. “SS. Antonio e Biagio e Cesare Arrigo (Tortona).
Before enrolling, all patients provided written informed consent.
This work was written according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement8.
Statistical analysis
Clinico-pathological variables were reported using descriptive analysis. Continuous variables were summarized with means and interquartile ranges (IQR) and compared with Wilcoxon test, while categorical variables were summarized using percentages and compared using chi-square or fisher test where appropriate.
Logistic regression was used to evaluate covariates that could potentially predict rCR.
All models were based on clinically relevant features such as menopausal status, hormone receptor status, HER2 expression (2 + FISH amplified vs 3 + ), metastatic sites defined as non-visceral (skin, soft tissues, lymph nodes, and breast), bone (with or without non-visceral metastases), and visceral (which included all the remaining sites), number of metastatic lesions, type of CT and anti-HER2 therapy received in the first-line setting, disease-free interval (DFI, defined as the time from the completion of any treatment for early BC until the identification of disease recurrence), and best response obtained throughout treatment.
Multivariable models were designed through stepwise selection after single imputation of missing values of menopausal status, hormone receptor status, HER2 status, number of metastatic sites, visceral and non-visceral metastases, encephalic involvement, type of anti-HER2 therapy, type of CT backbone, and DFI. Single imputation was performed assuming monotone missing patterns and using the logistic regression method.
The prognostic impact of the different variables in terms of overall survival (OS, defined as the time from metastatic diagnosis to death for any cause) was evaluated with the Kaplan Meyer method and compared using log-rank test.
Unadjusted and adjusted Cox models were used to compare survival outcomes among patients with or without rCR. Adjustments were made for known prognostic factors, including hormone receptor status, metastatic sites, number of metastatic lesions, and type of anti-HER2 therapy received in the first-line setting.
Results
Clinico-pathological characteristics of the entire patient cohort
At the time of data cut-off (July 11, 2022), 3423 patients with MBC had been enrolled in the GIM14 BIO-META trial, of which 814 presented HER2-positive disease. Data about the best radiological response were available for 545 patients, which were included in the final analysis (Fig. 1).
As summarized, 3423 patients were enrolled in the GIM14 BIO-META study, but after excluding those with HER2-negative MBC, as well as those treated with ET, those with TTD < 3 months and the ones with missing data regarding best radiological response, 545 patients were ultimately considered eligible for this analysis. MBC metastatic breast cancer, ET endocrine therapy, TTD time-to-treatment discontinuation.
The median follow up, computed using the reverse Kaplan–Meier method, was 6.06 years (IQR 3.38–9.50).
At MBC diagnosis, the median age was 53 years old (IQR 44–63) and most patients (333, 61%) were in post-menopause. Most cases (371, 68%) were hormone receptor-positive and presented an HER2 IHC score 3+ (324, 59%). With regard to the 371 patients with hormone receptor-positive HER2-positive MBC, most (270, 73%) did not receive maintenance endrocrine therapy (ET). Amongst the remaining 101 (27%) patients who did receive maintenance ET, aromatase inhibitors were the most commonly employed agents (64, 74%). Furthermore, 325 patients (60%) had recurrent MBC, and among them, 264 (81%) had received previous neoadjuvant or adjuvant treatment, mostly consisting of anthracyclines and taxanes (174, 66%). The median time from initial diagnosis of BC to radiological evidence of metastatic disease was 41 months (IQR 24.50-73.49). Additionally, most patients presented visceral metastases (314, 58%), 1 metastatic site (257, 47%) and no central nervous system (CNS) involvement (506, 93%). Taxanes were the main CT backbone (396, 73%) and 294 patients (54%) had received the trastuzumab-pertuzumab doublet.
Eighty patients (15%) experienced a rCR to first line anti-HER2 therapy, whilst the best response was a partial response (PR) for 250 patients (46%), stable disease (SD) for 181 patients (33%) and the remaining 34 patients (6%) experienced disease progression (PD)(Table 1).
The clinico-pathological characteristics of patients experiencing a rCR and of those with no rCR are summarized in Table 2.
Association amongst clinico-pathological variables and rCR
At univariable analysis, hormone receptor status (OR 1.72, p-value = 0.03), HER2 IHC score 3+ (OR 2.28, p-value < 0.01), first-line CT with anthracyclines and taxanes (OR 2.74, p-value = 0.01), presence of non-visceral metastases (OR 2.27, p-value < 0.01) and 1 metastatic site (OR 2.93, p-value = 0.04) were significantly associated with a higher probability of experiencing a CR with a TTD > 3 months (Supplementary Table 1).
However, at multivariable analysis, only HER2 IHC score 3+ (OR 2.03, p-value = 0.01), the presence of non-visceral metastases (OR 1.51, p-value = 0.01) and 1 metastatic site (OR 2.49, p-value = 0.01) were significantly associated with higher odds of obtaining a CR with a TTD > 3 months (Supplementary Table 1)(Fig. 2).
In detail, the three forest plots represented in this figure are referred to all patients with rCR (A), in patients with rCR and treated with trastuzumab-pertuzumab (B), and in patients with CR and TTD >18 months (C). rCR radiologic complete response, CR complete response, TTD time-to-treatment discontinuation.
Notably, of the 294 patients treated with the trastuzumab-pertuzumab doublet, 38 (13%) experienced a rCR. As the combination of a taxane and trastuzumab-pertuzumab is the current standard first-line treatment for patients with HER2-positive MBC, a sensitivity analysis focused on the clinico-pathological characteristics of this category of patients was performed and is reported in Supplementary Table 2 and Fig. 2.
Interestingly, of the 80 patients achieving a rCR, 56 (73%) experienced a durable and sustained CR with a TTD > 18 months, while 21 (27%) patients had a CR with a TTD < 18 months. Also, another 3 patients had a CR with an observation time inferior to 18 months, thus they were excluded from the following sub-analysis.
Subpopulation with sustained response (TTD > 18 months)
Amongst the 545 patients included in the analysis, 235 (43%) presented a sustained response with a TTD > 18 months. Concerning the best radiological response reported for such long responders, 56 patients (24%) achieved a CR, 96 patients (41%) experienced a PR and 83 (35%) had a stable disease, none experienced a disease progression (Supplementary Table 3). On the contrary, amongst the 281 patients (52%) with a TTD < 18 months, 21 (8%) presented a CR, 139 (49%) had a PR, 87 (31%) had a SD and 34 (12%) experienced a disease progression (Supplementary Table 3). For the remaining 29 patients, this information was incomplete.
Considering the subgroup of patients with such sustained CR, most (38, 68%) were post-menopausal at diagnosis, they mostly presented visceral metastasis (28, 50%), 1 metastatic site (37, 66%) and no CNS involvement (53, 95%) (Supplementary Table 4). Twenty-eight of these patients (50%) had hormone receptor-negative disease and most (44, 79%) presented an HER2 IHC score 3+ (Supplementary Table 4). Taxanes were the main CT backbone employed also in this subgroup of patients (41, 73%) and they mostly received the trastuzumab-pertuzumab doublet (33, 58%) (Supplementary Table 4).
The clinico-pathological characteristics of patients with radiological evidence of PR or SD and TTD > 18 months are reported in Supplementary Table 5.
At univariable analysis, hormone receptor-negative status (OR 3.20, p-value = 0.04), first-line CT with agents other than a taxane (OR 0.33, p-value = 0.04) and type of anti-HER2 therapy (OR 0.10, p-value = 0.04 for trastuzumab-based therapy and OR 0.14, p-value < 0.01 for other anti-HER2 therapy) were significantly associated with a higher probability of experiencing a CR with a TTD > 18 months (Supplementary Table 6).
Notably, at multivariable analysis, only type of anti-HER2 therapy (OR 0.13 p-value < 0.01 for trastuzumab-based therapy and OR 0.08, p-value = 0.04 for other anti-HER2 therapy) was found to be related to a higher probability of achieving a CR with a TTD > 18 months (Supplementary Table 6) (Fig. 2).
Survival outcomes
Considering all 545 patients with TTD > 3 months, a median OS of 10.22 years (IQR 8.21–12.73) was observed for those that had achieved a CR to a first-line anti-HER2 therapy. Instead, median OS was 6.71 years (IQR 6.07-8.84) and 6.29 years (IQR 5.33–7.27) for patients achieving a PR or a SD, respectively (Fig. 3). Moreover, compared to patients without rCR, those with rCR had a lower risk of OS events (unadjusted HR 0.50, 95% CI 0.34-0.75, p-value < 0.001; adjusted HR 0.50, 95% CI 0.33–0.74, p-value < 0.001). Interestingly, considering only patients with a single metastatic site, which accounted for almost 50% of the analyzed cohort, those with rCR had a lower risk of OS events compared to those without rCR (unadjusted HR 0.41, 95% CI 0.24–0.69, p-value = 0.001; adjusted HR 0.40, 95% CI 0.24-0.69, p-value < 0.001).
MBC metastatic breast cancer, PD disease progression, CR complete response, PR partial response, SD stable disease.
Focusing on the 235 patients with a TTD > 18 months, a median OS of 12.73 years (IQR 8.93-not reached) was observed amongst patients experiencing a CR, whilst median OS was 9.35 years (IQR 6.29-not reached) and 8.56 years (IQR 6.58–15.37) for those achieving a PR or a SD, respectively (Fig. 4).
MBC metastatic breast cancer, TTD time-to-treatment discontinuation, CR complete response, PR partial response, SD stable disease.
More in detail, considering only the patients that experienced a sustained CR, at 36 months an OS rate of 98% [95% confidence interval (CI) 87.35–99.73] was observed. Instead, the OS rate at 60 months was 92% (95% CI 79.31–96.81).
Overall, of the 545 patients included in the present analysis, 407 had discontinued first-line anti-HER2 therapy at the time of data cut-off due to the occurrence of a TTD event. The TTD for patients with CR was 47.20 months (IQR 16.32–132.07). Among these 407 patients, 127 (31%) received second-line therapy with trastuzumab emtansine (T-DM1), 59 (15%) received trastuzumab in combination with a CT, 50 (12%) received lapatinib and capecitabine, 44 (11%) were treated with various regimens of CT or ET in combination with an anti-HER2 agent, 29 (7%) received only CT without an anti-HER2 agent, and for the remaining 98 (24%) patients, the information was not available. Due to the wide variability of second-line regimens employed, post-progression survival analyses were not performed.
Finally, as shown in Fig. 1, 269 patients were excluded from the present analysis (69 for having received first-line ET in combination with anti-HER2 therapy, 90 for having a TTD < 3 months, and 111 for missing data on the best radiological response). The clinico-pathological characteristics of all excluded patients (Supplementary Table 7) were similar to those of the included population, except for menopausal status (64% versus 61% postmenopausal patients in excluded and included groups, respectively) and DFI (30% versus 40% of patients with de novo MBC in the excluded and included groups, respectively). Furthermore, among the excluded patients, only 31% received first-line therapy with the trastuzumab-pertuzumab doublet, compared to 54% in the analyzed cohort. Median OS was significantly shorter in the excluded population compared to the analyzed cohort [54.44 months (IQR 22.02-115.03) versus 79.47 months (IQR 43.64–164.27), p-value < 0.01)], which could potentially be related to the lower proportion of patients treated with the trastuzumab-pertuzumab doublet in the excluded group. More in detail, when comparing the 111 patients excluded for missing data on the best radiological response achieved to the 545 patients included in the analysis, no major differences in the clinico-pathological characteristics were observed (Supplementary Table 8). Notably, a lower percentage of these patients were treated with pertuzumab-based regimens (40%) compared to included patients (54%), but no significant difference in median OS was reported among the two groups [63.74 months (IQR 35.43–139.04) versus 79.47 months, p-value = 0.07].
Discussion
The present study evaluated a cohort of patients with HER2-positive MBC treated with a first-line anti-HER2 therapy with the aim of determining the clinico-pathological characteristics predictive of a CR. The cohort of interest included a total of 545 patients, most of which had a hormone receptor-negative and HER2 IHC score 3+ disease, visceral metastases with mainly a single metastatic site and treated mostly with a taxane in association with the trastuzumab-pertuzumab doublet.
Among these patients, 80 (15% of the analyzed cohort) experienced a CR with a TTD > 3 months and 56 had a TTD > 18 months, a higher proportion as compared to what observed in current literature. In fact, the primary results of the CLEOPATRA trial, reported that 19 patients randomized in the experimental arm with pertuzumab (5.5%) achieved a CR, 256 (74.6%) had a PR, 50 (14.6%) had SD and another 13 patients (3.8%) experienced a disease progression4. Furthermore, a retrospective study by Yeo et al. of patients with HER2-positive MBC treated with trastuzumab and CT regimens in the first-line setting, reported an objective response rate (ORR) of 65%, of which 17 (8%) were CR and 120 (57%) were PR, with durable responses5.
The better clinical outcome observed in our experience could possibly be related to the fact that a higher proportion of patients (40%) had de novo MBC with respect to the experience reported by Yeo et al. (20%). Moreover, in the CLEOPATRA trial, up to 78% of patients presented visceral metastases, compared to 58% in our study. This finding, along with the higher proportion of oligometastatic patients included in our analysis, could explain the difference observed in the clinical outcomes.
Interestingly, our experience showed that HER2 IHC score 3+, non-visceral metastases and 1 metastatic site were all significantly associated with higher odds of experiencing an rCR. Furthermore, with regards to the long responding patients experiencing a sustained CR with a TTD > 18 months, the type of anti-HER2 therapy received was significantly associated with the probability of experiencing such clinical outcome. Similarly, in the final analysis of the CLEOPATRA trial, 99 patients treated with the trastuzumab-pertuzumab doublet, who were considered long-term responders, mostly presented non-measurable, non-visceral, progesterone receptor-positive disease and HER2 IHC score 3+9. Additionally, a retrospective study by Murthy et al. evaluated patients with HER2-positive MBC that received anti-HER2 therapy (mostly trastuzumab), reporting a 5-year and 10-year OS rate of 33% and 7%, respectively10. In this study, younger age at MBC diagnosis, hormone receptor-positive disease and the presence of a single metastatic site, were all correlated with longer survival10. Furthermore, the German HER-OS patient registry enrolled patients with HER2-positive MBC treated with first-line therapy with trastuzumab and, as reported by Witzel et al., younger age at MBC diagnosis and good performance status were associated with a significantly longer time-to-progression, whilst no impact was observed for hormone receptor status11. Durable disease control can also be achieved with trastuzumab monotherapy, which has been significantly associated with improved OS in the SAKK 22/99 trial, in which the presence of visceral metastases and hormone receptor-negative status were described as negative predictive factors6.
As mentioned, most of the patients evaluated in our study had HER2-positive disease defined as HER2 IHC score 3+. The prognostic and predictive impact of HER2 scoring at IHC has been evaluated in an ancillary analysis of the MARIANNE trial, which compared a first-line treatment for HER2-positive MBC with T-DM1 + placebo or T-DM1 + pertuzumab versus the standard therapy consisting of a taxane + trastuzumab at the time12. In this exploratory analysis, patients were divided in HER2 IHC score 2+ and HER2 IHC score 3+ and the staining at IHC were further categorized as focal, heterogeneous or homogeneous distribution according to the total amount of cells that stained with an intensity of 2+ or 3+12. Notably, a focal or heterogenous staining distribution was rarely observed in IHC 3+ samples (2.8 and 11.8% respectively), whilst a higher proportion of IHC 2+ samples exhibited a focal distribution of the staining (48.1%)12. A higher progression-free survival (PFS) was observed in patients with HER2 IHC score 3+ with respect to those with an IHC score 2+12. Furthermore, the homogeneous staining in both IHC score 3+ and 2+ was associated with longer a PFS compared to the focal or heterogenous staining, especially in the T-DM1 arms12.
With this regard, another exploratory analysis of the GIM14 BIO-META trial has been conducted and presented at the San Antonio Breast Cancer Symposium 2023. In this work, Dri et al. described how an HER2 IHC score 3+ is less likely associated with the presence of visceral metastases13. Furthermore, the presence of CNS metastases was associated with shorter OS in IHC score 3+ and a worse OS was reported in patients with liver metastases with both HER2 IHC score 3+ and 2+13.
We reported a median OS of 10.22 years for patients with a rCR and of 12.73 years for those with a sustained CR (TTD > 18 months). A favorable median OS was also observed among patients experiencing a PR (6.71 years and 9.35 years, respectively). Although clinically relevant, a sustained PR does not hold the same prognostic impact as achieving a CR. Furthermore, patients with a CR could potentially be evaluated in the future for de-escalation strategies, which would be less feasible for patients with persistent evidence of residual disease.
Genomic signatures and HER2DX have gained momentum for the management of patients with early breast cancer (EBC). In particular, in patients with hormone receptor-positive, HER2-positive EBC, HER2DX has been found to be predictive of response to neoadjuvant treatment with letrozole and trastuzumab-pertuzumab, thereby helping to select patients which could benefit from a de-escalation strategy in this setting14. In patients with MBC, these tools could be considered to predict the depth of response; however, the prospect of treatment de-escalation in patients with a sustained CR will likely rely on higher sensitivity assays and circulating techniques capable of evaluating minimal residual disease (MRD).
As mentioned, in light of the considerable benefit reported in PFS and OS, the standard first-line therapy for HER2-positive MBC, encompasses the combination of a taxane and the trastuzumab-pertuzumab doublet for 6-8 cycles, followed by a maintenance therapy with the trastuzumab-pertuzumab doublet until PD4,15,16,17. However, defining the optimal duration of therapy in this category of patients represents a challenge and emphasis is now growing with respect to the possibility of interrupting early the maintenance treatment in case of CR18. Interestingly, a recently published case series, described the cases of 4 patients with HER2-positive MBC that experienced a durable CR (ranging from 1 to 5 years) to first-line therapy and subsequently interrupted maintenance treatment, without no subsequent evidence of disease recurrence19.
In the aforementioned experience by Murthy et al., 2 patients with HER2-positive MBC interrupted treatment with trastuzumab after having achieved a CR after about 10 and 12 years of therapy, respectively, with no evidence of disease recurrence after more than 7 years of follow-up10.
On the contrary, in the study performed by Witzel et al., the interruption of trastuzumab maintenance therapy was associated with a shorter time-to-progression11.
Thereby, identifying the patients that could safely interrupt the anti-HER2 maintenance therapy represents a challenge. In this setting, the phase II STOP-HER2 trial (NCT05721248) will evaluate the safety of anti-HER2 treatment interruption in patients with HER2-positive MBC who are defined as exceptional responders (disease control for at least 3 years from the start of first-line treatment).
With this regard, liquid biopsy, through the identification of MRD, could serve as a turning point.
Although linked to an increasingly hot topic in current clinical practice, several limitations should be considered while interpreting our data. First, its retrospective design could have introduced unadjusted and unknown confounding factors. Furthermore, even though 111 patients were excluded due to missing data on the outcome of interest, no relevant differences in clinico-pathological characteristics or long-term outcomes (i.e. OS) were observed. Notably, included patients were more frequently treated with pertuzumab-based regimens suggesting a more contemporary cohort. Furthermore, our study included a heterogeneous cohort of patients enrolled from multiple institutions and treated with different types of anti-HER2 regimens in several historical phases. This may enhance on the one hand the transferability of the results, but on the other may dilute specificities linked to more modern treatment regimens. Additionally, almost 50% of the patients evaluated in the present analysis had a single metastatic site, and data regarding the potential association of systemic treatments with locoregional approaches (such as surgery, ablative therapies, or radiation therapy) were not available, not even for patients with brain metastases. Finally, the features linked to higher odds of rCR with trastuzumab-pertuzumab could not be applicable to patients treated with novel agents, including novel antibody drug conjugates such as trastuzumab deruxtecan, whose role as first-line treatment of HER2-positive MBC is currently being investigated in the DESTINY-Breast09 trial (NCT04784715).
In conclusion, in our study higher odds of achieving a complete response to a first-line anti-HER2 therapy with a time-to-treatment-discontinuation > 3 months were present in case of HER2 Immunohistochemistry score 3+, visceral metastases and involvement of a single metastatic site. Furthermore, a sustained complete response was observed in the subset of patients that had been exposed to a trastuzumab-based therapy.
Continuing the treatment until unacceptable toxicity or disease progression is the current standard of care in this setting, but de-escalation strategies represent a growing need for patients experiencing durable complete response, especially due to the significantly increasing efficacy of new anti-HER2 agents.
Our study represents a starting point to properly identify patients that could be selected for future studies focused on safely interrupting maintenance anti-HER2 therapy through high-sensitivity diagnostic approaches.
Responses