Introduction

Most TNBC are highly proliferative cancers that lack the expression of estrogen and progesterone receptors as well as amplification of the Her2 oncogene. [1]. Preoperative chemotherapy is the standard of care in TNBC because of the prognostic significance of the pathologic response on long-term outcome and the opportunity to tailor subsequent adjuvant therapy to the quality of the response obtained [3-5]. Several studies have indicated that patients with TNBC benefit from the addition of platinum to the neoadjuvant chemotherapy, however impact on survival is still uncertain [6-9]. Given the fact that platinum adds to the toxicity of the regimen, an important question is whether all TNBC do benefit to the same degree. Applying the HRDetect mutational-signature-based algorithm, fifty to sixty percent of TNBC harbour a HR repair deficiency explained by BRCA1/2 germline/somatic mutations or other genomic (germline and somatic) instabilities, which provides specific therapeutic opportunities for the use of DNA double-strand break-inducing agents, including platinum salts, anthracyclines, alkylating agents and poly-ADP-ribose polymerase (PARP) inhibitors [10]. Up to date,it is still debated whether patients with a homologous recombination deficient TNBC, more broadly defined than only BRCA1/2 mutations, benefit more than other TNBC patients from the addition of platinum. Some studies using an HRD-score, based on loss of heterozygosity, telomeric allelic imbalance and large-scale state transitions, support that this might be the case or at least that these patients respond better to neoadjuvant platinum combinations [11,12]. This hypothesis has not been investigated using germline mutation analysis of genes involved in DDR. By performing whole exome sequencing and investigating in more details the genes involved in the DDR machinery, we tried to identify genomic germline biomarkers allowing a better selection of the patients, who will benefit from therapies with DNA damaging agents such as platinum, in order to avoid useless toxicity.

Patient and Methods

Patient population

Between June 2015 and May 2016, 65 TNBC patients were included in a previously published phase II neoadjuvant BSMO 2014-01 study [13]. The prospective, multicenter phase II trial explored the efficacy of neoadjuvant dose-dense weekly paclitaxel and carboplatin, followed by biweekly epirubicin and cyclophosphamide. The primary objective was to determine the pCR rate. One of the preplanned secondary objectives was to examine the correlation between pCR and germline carrier status of mutations in DDR or HR genes. Patients older than 18 years and with operable stage II and III were included after signing an informed consent. Two patients were excluded from the analysis because they were not assessable for the primary endpoint (one received doxorubicin instead of epirubicin and one refused surgery). As reported in the publication of the clinical results, 20 extra patients were recruited and treated with the same regimen with the purpose to have 60 samples to do whole exome sequencing. Eleven patients (patient ID 0202; 0909; 0912; 1305; 1901; 1902; 2001; 2002; 2602; 2904; 3001) from the first publication did not consent for translational research and 8 consecutive patients(patient ID: 0107; 0913; 0205; 0206; 2104; 2105; 2610; 3004) from the extra pool agreed to participate to the genomic analysis. Triple negativity was defined as estrogen and progesterone receptor expression in less than 10% of tumor cells and no Her2 amplification as defined by Her2 IHC 0-1 or FISH ratio less than two (ASCO/CAP guideline recommendations for HER2 testing) [14].

Study procedures

All patients were treated for 12 weeks with weekly paclitaxel (wP) 80mg/m² concurrent with weekly carboplatin (Cp) at an area under the curve (AUC) dose of 2, followed by epirubicin 90mg/m² and cyclophosphamide 600mg/m² (EC) biweekly for four cycles with myeloid growth factor support on day 2. Response assessment was planned at 2 time points of the neoadjuvant systemic therapy. The extent of surgery and subsequent irradiation was performed according to the local guidelines of the participating centers and no further adjuvant chemotherapy was foreseen in the study, although this was at the discretion of the investigator. Subsequently, patients were prospectively followed for recurrence and survival status.

Pathological evaluation

Histopathologic evaluation of response after neoadjuvant chemotherapy was done in accordance to the Pinder tumour response system [15]. Pathologic response was determined locally without central pathologic review. All surgical pathology reports were centrally reviewed. pCR rate was defined as no remaining invasive cancer in the breast and resected axillary lymph nodes (ypT0/isypN0).

Germline BRCA1/2 testing

Germline testing for BRCA1/2 was performed in the individual institutions utilizing available validated gene panel tests, according national guidelines. Patients with a germline BRCA1/2 mutation or other breast cancer predisposition genes were counselled as per institutional guidelines and considering the gene risk profile and familial cancer phenotype.

Whole exome sequencing

Blood samples (EDTA) were obtained at diagnosis, before any treatment from 60 TNBC patients participating in the fore mentioned clinical study. DNA was extracted and sent to BGI (www.bgi.com) for genome sequencing. Whole exome sequencing (WES) was performed using the SureSelect Human All Exon V6 kit from Agilent for target enrichment. Paired-end sequencing was performed on an Illumina instrument. The lists of genomic variants (compared to the reference genome hg19) obtained for each patient were provided as Variant Call Format (VCF) files. Further filtering was performed to retain only variants strongly affecting protein structure (nonsense, frameshift and splice-site variants) and variants with a minor allele frequency (MAF) <0.01. To exclude false positives, variants occurring in 10% (or more) of the samples were also discarded. To categorize genes as participating in the DDR or HR pathway, we refer to a recent publication by Knijnenburg et al [16].The publication qualifies 276 genes as belonging to the DDR pathway, of which 88 belongs to the HR pathway and 21 to the HR core pathway. Data available in Table 6 All reported variants were manually reviewed and validated making use of the Integrative Genomic Viewer (IGV) from the Broad Institute [17]. All patients signed an informed consent allowing germline exome sequencing and the study was approved by the institutional ethics committee.

Statistical methods

The correlation between pCR rate and germline defects in genes involved in DDR or specifically in HR, was examined using the Fisher's exact test. All reported P values are from one sided tests for pCR correlation and two sided tests for correlation with hematologic toxicity. All analyses were performed using SPSS Statistics version 27.

Results

Patient characteristics

The demographics and clinicopathologic data of the sixty evaluable patients are shown in Table 1. Most of the patients were between 40 and 60 years old, with a median age of 55 (33-76yrs; SD: 11.7). Ninety percent of the patients had stage IIA or IIB disease, with a majority of T2 tumours and clinically node negative disease. Ninety seven percent of the patients were diagnosed with an invasive ductal carcinoma and in a large majority grade 3. One patient had a lobular carcinoma and one a mixed ductal and lobular carcinoma. Mastectomy was performed in 17 patients, breast conserving surgery in 43 and axillary dissection in 36 patients.

Table 1: Patient Characteristics (n=60)

Germline gene testing

Routine diagnostic germline gene panel testing was performed in an initial step and revealed a deleterious BRCA1/2 mutation in nine patients (15%). Subsequent whole exome sequencing (see Table 2 for an overview of the relevant data) could detect a germline DDR gene mutation in 35 (58%) of the 60 TNBC patients. In 19 of these 35 patients (31.5%) the mutated DDR gene was a HR gene. More specifically, fifteen of these 19 patients had a mutation in a HR core gene, including the nine patients with a BRCA1 or BRCA2 mutation and six patients with a deleterious mutation in the RAD50, RAD52, BARD1 or EME1 genes. One patient had a mutation in two different HR core genes (RAD50 and RAD52). Four patients had a mutation in non-core HR genes: RECQL5, RECQL4 and EME2. Sixteen patients had germline DDR mutations not involving the HR machinery (eg in CHEK2). All pathogenic BRCA1 and BRCA2 mutations found during routine testing were also detected in the subsequent WES. No additional BRCA1/2 mutations were identified. Each of the nine BRCA1 or BRCA2 mutations was found only once, whereas two different RAD52 mutations were found each in two patients.

Table 2: List of the pathogenic variants identified in DDR genes of 60 TNBC patients

Correlation between germline defects and pathologic complete response:

Association of DDR versus HR gene mutation and response to platinum-based chemotherapy

A pathologic complete remission was obtained in 68.5% of the DDR mutated patients compared to 40% in the non-DDR mutated population (p=0.026). When we restricted our analyses to the patients with a mutation in the HR genes, the pCR rates increased to 84% in comparison to 44% to the patients lacking a HR gene mutation (p = 0.003). When further considering only the HR core genes, a pCR was observed in 93% (14/15) of the patients and in 44% (20/45) of the patients without a HR core gene mutation (p = 0.0007).The only patient not presenting a pCR in this subgroup carried a BRCA1 mutation. For the patients with a DDR gene mutation not included in the HR core gene panel, the pCR dropped to 50% (10/20), which was much closer to what we found in patients without a DDR mutation (40%) as shown in Table 3 and Table 4.

Table 3: Correlation between germline defects in DDR genes of 60 TNBC patients and 5 clinical parameters

This table provides in the first column, the number of TNBC patients for which a germline mutation was found respectively in a DDR gene, in a DDR minus HR core gene, without a DDR germline mutation, in a HR gene, a HR core gene or the BRCA 1/2 genes. In the five subsequent columns the number of patients presenting the characteristics specific for each of five different clinical parameters are indicated. Statistical correlations between mutation carriership and each of the five clinical parameters were investigated using Fisher’s Exact tests.

Table 4: List of pathogenic variants in DDR genes of 60 TNBC patients and pathologic response.

Association of DDR versus HR gene mutation and hematologic toxicities.

DDR neither HR germline gene mutations did clearly predict for hematologic toxicities, such as febrile neutropenia G3 and G4 (p=0.78; p=1), neutropenia G3 and G4 (p=0.05; p=1), anemia G3 (p=1; p=0.76), trombopenia G3 and G4 (p = 0.73; p=0.26) as shown in Table 3 and Table 5. Since neutropenia G3 and G4 appeared to occur somewhat less frequently in patients without a DDR mutation (13/25) than in patients with such mutation (27/35; p=0.055, which is at limit of significance), we further compared patients without a DDR mutation to patients with a DDR mutation that did not benefit well from the therapy (excluding the patients with a HR core gene mutation). Patients with such a mutation suffered clearly more from neutropenia G3 and G4 (13/25) than patients without a DDR mutation (13/25; P=0.027, see Table 3).

Table 5: List of pathogenic variants and hematologic toxicities.

Table 6: List of 276 DDR genes, 88 HR genes and 21 HR core genes to be sequenced.

Discussion

Triple-negative breast cancer is generally more sensitive to neoadjuvant platinum-based chemotherapy than other subtypes of breast cancer [18]. Several reports indicated an increased effectiveness of preoperative platinum based systemic treatment in BRCA1/2 mutant and non-BRCA mutant HRD-positive breast cancer [11, 12, 18, 19]. In these studies, BRCA1/2 sequencing with or without an HRD-scoring was used [where the HRD score is the sum of three metrics of chromosomal level aberration: LOH (loss of heterozygosity), TAI (telomeric allelic imbalance) and LST (large-scale state transitions)]. It is so far unclear whether BRCA1/2 mutations or HRD as defined by an HRD score select for patients that benefit more from the addition of platinum. There are also no publications that examined the correlation of broad germline analysis of DDR or HR genes with response to neoadjuvant chemotherapy. In the current study we investigated to what extent a dose-dense platinum containing regimen was more efficacious in triple-negative breast cancer patients with or without a germline defect in the DNA repair machinery as defined from the sequencing of a panel of 276 DDR genes. Finding a predictive biomarker is essential as the inclusion of platinum significantly adds to the toxicity of the chemotherapy. On the other hand, striving for the highest efficacy is important as obtaining a pCR has crucial prognostic significance in terms of the risk of relapse and survival [4, 5]. We used whole-exome sequencing to maximize the discriminative power between cancer patients that have no DNA repair defect and those that do. Therefore, in this study the cohort without a germline DNA damage repair defect is less likely to be diluted by non-identified repair defects. For the design of the virtual DDR gene panel, we relied on a recent publication by Knijnenburg et al [16].The proportion of TNBC patients with a germline HR defect as defined in our study is 31.5 %, (25% if considering only HR core genes). The proportion of patients with any DDR germline defect was 58%. This means that a large fraction of TNBC patients have a proven or probable genetic etiologic factor. We did find a high pCR rate in the overall population (53%) consistent with other studies. A higher pCR rate was observed in patients with a DDR gene germline defect (68.5%). This increased pCR rate was driven by the patients with HR gene mutations (pCR rate of 84%), and more specifically by the patients with HR core gene defects (pCR rate of 93%). The pCR rate observed in the patients without a HR core gene defect was 44,4% and within the range of pCR rates found when platinum is not included in the neoadjuvant chemotherapy [20]. DDR defects other than HRD have also been proposed to sensitize for DNA damaging chemotherapy including cisplatinum. Our study does not support this as the pCR rates observed in patients without a germline DDR mutation and patients with a DDR gene mutation not including a HR core gene mutation are in the same range: 10/25 (40%) versus 20/45 (44.4%) respectively.Therefore, a logical proposal would be to restrict the addition of platinum to TNBC patients having a germline HR core gene mutation. It is clear that the current prospective phase 2 study does not prove that platinum is needed to achieve this result, and it could be that these tumours are simply more chemo-responsive cancers. However, our results are in line with other studies that show a high pCR rate in patients with BRCAness, which was defined in different manners [11, 12]. The GEPARsixto trial showed a significant benefit only in HRD triple negative breast cancers, although the authors did not consider these data as definitive because of the cohort size and there was no cyclophophamide included [12]. Comprehensive gene sequencing of the germline as in this study might enrich the specific HRD population and the discriminative power of the studies. A recent meta-analysis of Chai Y et al found a significant higher efficacy of platinum-based regimens in BRCA-mutated TNBC compared to BRCA wild type patients (p =0.002) and the same was true for the HRD-positive versus HRD-negative tumors with a p < 0.001 [21]. Another meta-analysis comparing platinum-based versus platinum-free neoadjuvant chemotherapy in TNBC patients showed no significant increase in pCR rate with the addition of platinum in the BRCA-mutated patients. In this last study, the authors underlined that the number of included BRCA patients was too small to correctly evaluate the effect of platinum compounds in mutated versus non-mutated patients [22].These studies have less than optimal discriminative power as the non-BRCA1/2 mutant cohorts also include patients with other HRD defects. Also the predictive value of HRD remains controversial as in the TNT phase 3 clinical trial there was no correlation between carboplatin response and a high score in a Myriad HRD assay [23]. Moreover only a HRD score threshold of 42 had the potential to identify patients who might benefit from platinum based preoperative systemic therapy and the HRD status is more suitable for variation between the groups.

The ongoing PEARL phase 3 trial comparing anthracyclines followed by a taxane with anthracyclines followed by taxane plus carboplatin as neoadjuvant treatment in TNBC patients, stratified by BRCA 1/2 mutation status will be published in 2023, but this study will have the same limitation of incomplete HRD specification.

Conclusion

The present study using a comprehensive germline DDR gene mutation analysis is hypothesis generating, and suggest that platinum addition in the neoadjuvant treatment of TNBC could be restricted to patients with a germline HR core gene mutation. However prospective phase 3 trials should include broad HR gene characterization beyond BRCA1/2 to confirm our finding. This study also indicated that a high proportion of TNBC has a possible genetic etiology. Finally, in accordance with these data, additional studies should be performed to investigate whether a clear correlation between germline mutation carriership in a HR core gene and a response to therapies targeting cells with double strand DNA break repair deficiency can also be found in other cancer types, such as prostate cancer.

Acknowledgements

We want to acknowledge all the co-investigators of the BSMO, the pathology department of UZ Brussel, the Laboratory of Medical and Molecular Oncology and the patients.

Funding

The study was financially supported by Amgen and Teva

Compliance with ethical standards

Conflict of interest

All authors declare to have no conflict of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

References

1. Boyle P. Triple-negative breast cancer: epidemiological considerations and recommendations. Ann Oncol 23 (2012): 7-12
2. Burnstein HJ, Curigliano G, Loibl S, et al. Estimating the benefit of therapy for early-stage breast cancer: the St. Gallen International Consensus Guidelines for the primary therapy of early-stage breast cancer. Ann Oncol 30 (2019): 1541-1557
3. Cardoso F, Kyriakides S, Ohno S, et al. Early breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 30 (2019): 1194-1220.
4. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and longterm clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet 384 (2014): 1664-1672
5. Huang M, O’Shaughnessy J, Zhao J, et al. Association of pathological complete response with long-term survival outcomes in triple-negative breast cancer: a meta-analysis. Cancer Res 80 (2020): 5427-5434
6. Loibl S, o'Shaughnessy J, Untch M, et al. Addition of the PARP inhibitor veliparib plus carboplatin or carboplatin alone to standard neoadjuvant chemotherapy in triple-negative breast cancer (BrighTNess): a randomized, phase 3 trial. Lancet Oncol 19 (2018): 497-509
7. Sikov WM, Berry DA, Perou CM, et al. Impact of the addition of carboplatin and/or bevacizumab to neoadjuvant once-per-week paclitaxel followed by dose-dense doxorubicin and cyclophosphamide on pathologic complete response rates in stage II to III triple negative breast cancer. CALGB 40603 (Alliance). J Clin Oncol 33 (2015): 13-21
8. von Minckwitz G, Schneeweiss A, Loibl S, et al. Neoadjuvant carboplatin in patients with triple negative and Her2-positive early breast cancer (GeparSixto; GBG 66): A randomized phase 2 trial. Lancet Oncol 15 (2014): 747-756
9. Sharma P, Lopez-Tarruella S, Garcia-Saenz JA, et al. Pathological response and survival in triple-negative breast cancer following neoadjuvant carboplatin plus docetaxel. Clin Cancer Res 24 (2018): 5820-5829.
10. Staaf J, Glodzik D, Bosch A, et al. Whole-genome sequencing of triple-negative breast cancers in a population-based clinical study. Nature Medicine 25 (2019): 1526-1533.
11. Telli ML, Timms KM, Reid J, et al. Homologous recombination deficiency (HRD) score predicts response to platinum-containing neoadjuvant chemotherapy in patients with triple-negative breast cancer. Clin Cancer Res 22 (2016): 3764-3773.
12. Loibl S, Weber KE, Timms KM, et al. Survival analysis of carboplatin added to an anthracycline/taxane-based neoadjuvant chemotherapy and HRD score as predictor of response-final results GeparSixto. Ann Oncol 29 (2018): 2341-2347.
13. Fontaine C, Renard V, Van den Bulk H, et al. Weekly carboplatin plus neoadjuvant anthracycline-taxane-based regimen in early triple-negative breast cancer: a prospective phase II trial by the Breast Cancer Task Force of the Belgian Society of Medical Oncology (BSMO). Breast Cancer ResTreat 3 (2019): 607-615.
14. Tchrakian N, Flanagan L, Harford J, et al. New ASCO/CAP guideline recommendations for Her2 testing increase the proportion of reflex in situ hybridization tests and of her2 positive breast cancers. Virchows 468 (2016): 207-211.
15. Pinder SE, Provenzano E, Earl H, et al. Laboratory handling and histology reporting of breast specimensfrom patients who received neoadjuvant chemotherapy. Histopathology 50 (2007): 409-417.
16. Knijnenburg TA, Wang L, Zimmerman MT, et al. Genomic and molecular landscape of DNA damage repair deficiency across the cancer genome atlas. Cell Rep 23 (2018): 239-254.
17. Peck B, Bland P, Mavrommati I, et al. 3D functional genomics screens identify CREBBP as a targetable driver in aggressive triple-negative breast cancer. Cancer Res 81 (2021): 847-859.
18. Von Minckwitz G, Fontanella C, et al. Selecting the neoadjuvant treatment by molecular subtype: How to maximize the benefit? The Breast 22 (2013): S149-S151.
19. Telli ML, Hellyer J, Audeh W, et al. Homologous recombination deficiency (HRD) status predicts response to standard neoadjuvant chemotherapy in patients with triple negative or BRCA 1/2 mutation-associated breast cancer. Breast Canc Res Treat 168 (2018): 625-630
20. Liedtke C, Mazouni C, Hess K, et al. Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol 26 (2008): 1275-1281.
21. Chai Y, Chen Y, Zhang D, et al. Homlogous recombination deficiency (HRD) and BRCA 1/2 gene mutation for predicting the effect of platinum-based neoadjuvant chemotherapy of early-stage triple-negative breast cancer (TNBC): a systematic review and meta-analysis. J Pers Med 12 (2022): 1-13.
22. Poggio F, Bruzzone M, Ceppi M, et al. Platinum-based neoadjuvant chemotherapy in triple-negative breast cancer: a systemic review and meta-analysis. Ann Oncol 29 (2018): 1479-1508
23. Tutt A, Tovey H, Cheang M, et al. Carboplatin in BRCA 1/2-mutated and triple-negative breast cancer BRCAness subgroups: The TNT Trial. Nat Med 24 (2018): 628-637.