Molecular Profile and Matched Targeted Therapy for Advanced Breast Cancer Patients

Round 1

Reviewer 1 Report

The paper consider the usefulness of NGS testing in assigning targeted therapies. The result shows that in 14% of patients it is possible. This is a typical value, found also for other cancers, but it's low enough to discourage doctors from using NGS, as discussed in the paper. The reviewer therefore wonders whether it would be possible to preselect patients from the overal pool of BC patients to the pool, which could yield a larger accrual ratio? Did the authors try to correlate the patient's attributes, available in hospital data with the two classes of patients? (i.e. the class of 5 patients who are eligible for target therapy, and the class of remaining patients -- is there any classical parameter, describing the medical/physiological state of a patient that is significantly different in the selected class?)
Except for this research question - the reviewer also would suggest to extend the introduction. The paper was difficult to read for a person from outside the NGS world - to understand the paper I had to search for the physiopathological  role of the investigated proteins and criteria for selection to a targeted therapy. Sketch of some description in this regard would greatly enhance the readibility of this paper for sciencists working in neighbouring research areas.

Author Response

Dear Editor,

Thanks for the opportunity to review the manuscript and the kind advice.

Please find below the answers point to point.

The paper consider the usefulness of NGS testing in assigning targeted therapies. The result shows that in 14% of patients it is possible. This is a typical value, found also for other cancers, but it's low enough to discourage doctors from using NGS, as discussed in the paper.

The reviewer therefore wonders whether it would be possible to preselect patients from the overal pool of BC patients to the pool, which could yield a larger accrual ratio? Did the authors try to correlate the patient's attributes, available in hospital data with the two classes of patients? (i.e. the class of 5 patients who are eligible for target therapy, and the class of remaining patients -- is there any classical parameter, describing the medical/physiological state of a patient that is significantly different in the selected class?)

Thank your for your suggestion. We added these data in results section and commented them in discussion.

There were no clinical characteristics significantly different among the group of pa-tients eligible for target therapy (G1, n=5) and the others (G2, n=30). Molecular subtypes of BC were homogeneously distributed among the two groups, expect for HER2-enriched subtype that was represented only in G1 (1/5=20%). Liquid biopsy and solid tissues were equally used in the two groups (liquid biopsy 20%, solid tissues 80%) for NGS analysis. About the genomic findings, patients in G1 had a higher number of variants of uncertain significance (VUS) compared to G2 (25.25 vs 11, p=0.0009).

[…] With the limit of the small sample size, we found that patients belonging to G1, that are those eligible to target therapies, had a significantly higher number of VUS compared to G2 but, interestingly, they do not differ for number of pathogenic mutations. A VUS is a change in the sequence of a gene which cannot be determined pathogenic (something that causes disease) or benign (something that does not cause disease). The effect of a VUS, whether somatic or germline, remains unclear and its clinical relevance is uncertain. We do not know if there is a correlation between the number of VUS and the value of TMB. Moreover, data about genes carrying a VUS should be collected to provide some correlations. For instance, a recent study showed that, across diverse cancers, VUS in POL (DNA polymerase) genes exhibited an additive effect as carriers of multiple VUS and had exponentially increased TMB and prolonged overall survival [22].

Except for this research question - the reviewer also would suggest to extend the introduction. The paper was difficult to read for a person from outside the NGS world - to understand the paper I had to search for the physiopathological  role of the investigated proteins and criteria for selection to a targeted therapy. Sketch of some description in this regard would greatly enhance the readibility of this paper for sciencists working in neighbouring research areas.

We extended the introduction, as for your kind advice.  

Reviewer 2 Report

The research article ‘Molecular profile and matched targeted therapy for advanced breast cancer patients’ submitted by Falcone R et al., presents a clinical study where authors try to evaluate the percentage of breast cancer (BC) patients assigned to precision therapy based on genomic findings by new generation sequencing (NGS). The study was carried out over 20 months and a total of 38 patients participated in this study. Authors found the prevalent genes altered in their genomic analysis were TP53, PIK3CA, 109 BRCA2, ESR1 and RAD21. Authors concluded a low accrual rate for BC patients and only 5 patients (14%) out of 38 received specific treatments based on their molecular analysis through NGS. I have a few concerns with this clinical study.

1.     Authors should elaborate on the introduction about the complexity of BC disease and how a specialized treatment could affect the survival of these patients.

2.     Could author discuss the factors that HR⁺/HER2^- and TNBC were the most frequent subtypes referred for NGS?

 

3.     TP53 was the most common genetic aberration found in the NGS analysis, authors should discuss it and a potential factor (s) behind this finding. Was there a correlation between the organ (tissue or blood) used for NGS and genetic aberrations associated with it?

Author Response

Dear Editor and Reviewers,

Thanks for the opportunity to review the manuscript and the kind advice.

We extended the manuscript.  Please find below the answers point to point.

The research article ‘Molecular profile and matched targeted therapy for advanced breast cancer patients’ submitted by Falcone R et al., presents a clinical study where authors try to evaluate the percentage of breast cancer (BC) patients assigned to precision therapy based on genomic findings by new generation sequencing (NGS). The study was carried out over 20 months and a total of 38 patients participated in this study. Authors found the prevalent genes altered in their genomic analysis were TP53, PIK3CA, 109 BRCA2, ESR1 and RAD21. Authors concluded a low accrual rate for BC patients and only 5 patients (14%) out of 38 received specific treatments based on their molecular analysis through NGS. I have a few concerns with this clinical study.

1. Authors should elaborate on the introduction about the complexity of BC disease and how a specialized treatment could affect the survival of these patients.

We modified the Introduction section, as for your suggestion.

Breast cancer (BC) is a complex and heterogeneous disease [1], including several subtypes differing for clinical behavior, prognosis and response to therapies. The traditional molecular classification of BC (luminal A, luminal B, triple negative, her2 enriched) has been incorporated into clinical care practice [2]. On the contrary, the genomic landscape of advanced disease is an emerging issue investigated by next-generation sequencing (NGS) technologies. Some researchers combined genomic data with clinical details in the effort to identify those molecular alternations that mediate progression or drug resistance and to support targeted therapy decisions [3]. Indeed, tumor sequencing results can guide clinical trial enrolment or identify drug opportunities in individual patients. Detailed analysis of the genome of BC offers the chance to personalise therapy with several limitations: availability of targeted agents, interpretation of genomic alterations and their roles (drivers versus passengers), agreeing on rules to prioritize actionable findings. Moreover, the spatial and temporal tumour evolution represents an important clinical issue [4]. Archival primary tumour tissue cannot be assumed to be representative of the advanced disease genomic profile and poses a clinical challenge in regard to acquisition of multiple longitudinal tissue biopsies [5].

[…] Moreover, the portfolio of experimental therapies available in each hospital and the ability to perform the analyses in clinically acceptable time are two critical points to address.

Over the past 20 years, new classes of agents (CDK 4/6 inhibitors, antibody drug conjugate, PARP inhibitors, immunotherapy) significantly improved survival outcomes of BC patients [11, 12] and offered new treatment opportunities for those progressing to standard therapies. Recently, trastuzumab-deruxtecan (a HER2 antibody-drug conjugate) showed impressive results among HER2+ advanced BC [13] with 76% of patients alive without disease progression at 12 months. Two PARP inhibitors, olaparib and tala-zoparib, have been approved for treatment of germline BRCA carriers with metastatic HER2 negative BC. They showed, compared to chemotherapy, improvement in progres-sion free survival (PFS) and quality of life but no difference in overall survival (OS) [14].

2. Could author discuss the factors that HR⁺/HER2^- and TNBC were the most frequent subtypes referred for NGS?

Many thanks for your advice. We discussed that in the discussion section.

Half of referred patients for NGS were HR⁺/HER2^- BC, with 39% of TNBC, showing as the need of new treatments for these subtypes is a priority. Although TNBC represents a minority of BC, the large use of NGS in our experience may demonstrate an effort by oncologists to find actionable genes in such a heterogeneous population with poor treatment options and poor outcomes. On the contrary, HR⁺/HER2^- advanced BC is a common disease with longest overall survival, better outcome and the chance to underwent multiple treatments over years [27]. In HR⁺/HER2^- disease, clinicians and patients may prefer targeted agents over chemotherapy to postpone chemotherapeutics as much as possible. Plasma-based testing were ordered equally in hormone receptor–positive subtype and TNBC: in the first case, the high frequency of bone metastases make difficult to get diagnostic tissue for genomic analysis. Among TNBC, physicians may favor genomic blood tests, whom results are available in shorter time.

3. TP53 was the most common genetic aberration found in the NGS analysis, authors should discuss it and a potential factor (s) behind this finding. Was there a correlation between the organ (tissue or blood) used for NGS and genetic aberrations associated with it?

Thanks for the suggestion.

We added some details in RESULTS section with table 2, and discussed them in the discussion section.

The 22 patients with TP53 mutations carried 25 TP53 mutations (see table 2 in the manuscript).

Genomic alterations in TP53 were prevalent among TNBC (50% of all TP53-mutant BC) and observed largely in solid tissue (72% vs 28% in blood). All patients (100%) with liver metastases had TP53 mutations. No differences among TP53+ and TP53- BC were identified for other metastatic sites.

[…] In particular, TP53 mutations are the most frequent genetic alterations in breast cancer, observed largely in TNBC [24]. Previous studies reported that TP53 mutations are significantly associated with shorter OS and are an independent predictive factor of OS for BC patients [25]. Although TP53 mutations were homogenously distributed throughout the metastatic sites, all patients with liver lesions had TP53 mutations. This finding is not supported by other experiences. Concordance rates for TP53 mutations between breast cancer tissue and plasma are extremely different across studies, which might be dependent on disease subtype, stage, time of sampling, technology used for NGS analysis and the VAF threshold [26].

Reviewer 3 Report

The work presented for review is interesting. The authors presented the results of molecular evaluation of biological material from 38 breast cancer patients. They used the NGS method for molecular evaluation. The study aimed to identify new and further targeted treatment options in metastatic or recurrent breast cancer. As the authors point out in the paper, only a small percentage of patients have NGS testing performed for new therapies. Lack of access to clinical trials for oncology patients is a common topic of discussion among physicians who are looking for new treatment options for patients. The authors analyzed 38 patients and only 5 patients benefited from clinical trials which is only 14%. Based on such a small group, it is difficult to make further conclusions. However, the work indicates that NGS analysis in breast cancer can translate into improved outcomes and longer survival times, which requires a larger cohort of patients. The paper is eligible for publication because it is further evidence that molecular analysis in oncology points to further treatment options for patients.

Author Response

Dear Editor and Reviewers,

Thanks for the opportunity to review the manuscript and the kind advice.

We extended the manuscript  and added further details. 

Round 2

Reviewer 2 Report

Authors have addressed my concerns.