Molecular profiling of metastatic breast cancer and target-based therapeutic matching in an Asian tertiary phase I oncology unit

Introduction Molecular profiling of metastatic breast cancer (MBC) through the widespread use of next-generation sequencing (NGS) has highlighted actionable mutations and driven trials of targeted therapy matched to tumour molecular profiles, with improved outcomes reported using such an approach. Here, we review NGS results and treatment outcomes for a cohort of Asian MBC patients in the phase I unit of a tertiary centre. Methods Patients with MBC referred to a phase I unit underwent NGS via Ion AmpliSeq Cancer Hotspot v2 (ACH v2, 2014–2017) prior to institutional change to FoundationOne CDx (FM1; 2017–2022). Patients were counselled on findings and enrolled on matched therapeutic trials, where available. Outcomes for all subsequent treatment events were recorded to data cut-off on January 31, 2022. Results A total of 215 patients were enrolled with successful NGS in 158 patients. The PI3K/AKT/PTEN pathway was the most altered with one or more of the pathway member genes PIK3/AKT/PTEN affected in 62% (98/158) patients and 43% of tumours harbouring a PIK3CA alteration. Tumour mutational burden (TMB) was reported in 96/109 FM1 sequenced patients, with a mean TMB of 5.04 mt/Mb and 13% (12/96) with TMB ≥ 10 mt/Mb. Treatment outcomes were evaluable in 105/158 patients, with a pooled total of 216 treatment events recorded. Matched treatment was administered in 47/216 (22%) events and associated with prolonged median progression-free survival (PFS) of 21.0 weeks [95% confidence interval (CI) 11.7, 26.0 weeks] versus 12.1 weeks (95% CI 10.0, 15.4 weeks) in unmatched, with hazard ratio (HR) for progression or death of 0.63 (95% CI 0.41, 0.97; p = 0.034). In the subgroup of PIK3/AKT/PTEN-altered MBC, the HR for progression or death was 0.57 (95% CI 0.35, 0.92; p = 0.02), favouring matched treatment. Per-patient overall survival (OS) analysis (n = 105) showed improved survival for patients receiving matched treatment versus unmatched, with median OS (mOS) of 30.1 versus 11.8 months, HR = 0.45 (95% CI 0.24, 0.84; p = 0.013). Objective response rate (ORR) in the overall population was similar in matched and unmatched treatment events (23.7% versus 17.2%, odds ratio of response 1.14 95% CI 0.50, 2.62; p = 0.75). Conclusions Broad-panel NGS in MBC is feasible, allowing therapeutic matching, which was associated with improvements in PFS and OS.


Introduction
Precision oncology has drastically changed treatment paradigms in many solid organ tumours over the last decade, driven by an exponential increase in our knowledge of somatic molecular aberrations and the ability to target key oncogenic drivers.Knowledge of such aberrations is facilitated by the widespread adoption of next-generation sequencing (NGS) for somatic profiling, allowing molecularly targeted, or "matched", therapies to be administered to patients more frequently.Precision oncology has seen great success in non-small cell lung carcinoma (NSCLC), with current National Comprehensive Cancer Network (NCCN) and European Society for Medical Oncology (ESMO) Precision Medicine Working Group guidelines recommending testing for actionable alterations via a broad-panel approach in newly diagnosed advanced NSCLC (1, 2).
Metastatic breast cancer (MBC) has traditionally had limited utility for broad-panel somatic molecular profiling, with treatment decisions driven by hormone receptor status and the presence or absence of CerbB2 overexpression/amplification.Poly (ADPribose) polymerase (PARP) inhibitors olaparib and talazoparib improve progression-free survival (PFS) over standard-of-care chemotherapy in germline BRCA1/2-altered human epidermal growth factor receptor 2 (HER2)-negative MBC, which accounts for approximately 10% of MBC cases (13,14).Both agents received subsequent approval; however, to date, this does not extend to their use in patients with somatic BRCA1/2 mutations alone, though early data have suggested a potential role in this context (15).
Multiple trials have attempted to target mutations in PIK3CA, the gene encoding phosphatidylinositol 3-kinase (PI3K) catalytic subunit, which is altered in approximately 40% of MBC.Most notably in 2019, the phase III SOLAR 1 trial reported that alpelisib in combination with fulvestrant led to improved PFS over fulvestrant alone in hormone receptor positive, HER2-negative MBC harbouring hotspot mutations in PIK3CA, leading to the subsequent approval for this combination (16).Further studies have looked to target PI3K pathway alterations beyond PIK3CA.Both the phase II FAKTION and phase III CAPItello-291 trials have examined the addition of capivasertib, an oral inhibitor of AKT, to fulvestrant in patients with oestrogen receptor (ER)-positive, HER2-negative MBC after progression or relapse on an aromatase inhibitor (17,18).The FAKTION authors reported a significant PFS and overall survival (OS) benefit favouring the novel combination; however, subgroup analysis showed that this benefit was primarily in patients considered to have PIK3/AKT/PTEN pathway alteration, with no significant benefit seen in non-altered tumours (17).Results of CAPItello-291 support a significant PFS benefit in the overall patient population with the addition of capivasertib to fulvestrant, with HR for progression or death of 0.60 [95% confidence interval (CI) 0.51, 0.71], although the magnitude of benefit was greater in those with AKT pathway alterations, HR = 0.50 (95% CI 0.38, 0.65) (18).
Phase I studies have provided evidence that molecularly matched therapy can lead to a similar benefit in refractory MBC, with several non-randomised studies showing the administration of matched therapy to be associated with improved outcomes (19)(20)(21).In a retrospective analysis including only MBC (n = 97), patients on matched clinical trials had improved PFS (HR = 0.52, p = 0.003) and OS (HR = 0.54, p < 0.001) versus those on non-matched trials (19).
Here, we report a retrospective analysis of MBC patients enrolled in the Integrated Molecular Analysis of Cancer (IMAC) study from April 28, 2014, to January 31, 2022.The IMAC study is an ongoing prospective trial using broad-panel sequencing of refractory solid-organ malignancies to identify targetable molecular alterations in the Phase I unit of the National University Cancer Institute, Singapore (NCIS).Earlier results detailing the feasibility and outcomes from April 28, 2014, to September 1, 2016, have been reported (22).We present broadpanel sequencing results as well as clinical outcomes of future treatment lines after administration of molecularly matched or unmatched therapy.

Patient selection
Patients with MBC enrolled in the IMAC program from April 28, 2014, to January 31, 2022, were included in the current retrospective analysis.The IMAC study enrolled patients with advanced solid organ malignancies reviewed in the Developmental Therapeutic Unit of NCIS.Available tumour tissue was obtained for NGS based on paraffin-embedded primary tumour and/or metastatic specimens, which were obtained during routine clinical care (e.g., surgery, biopsy or ascites, and pleural fluid drain).Wherever possible, fresh tumour samples were collected, with archival tumours utilised when a fresh sample was not available or technically feasible to obtain.Additionally, commencing from 2017, blood samples for circulating tumour (ct) DNA analysis were eligible for testing at the discretion of the patient's primary physician should there be insufficient tumour tissue.Patients with successful sequencing were included in the mutational profile analysis of this current retrospective report.

Sequencing and analysis
For patients enrolled from April 2014 to July 2017, sequencing was via amplicon-based Ion AmpliSeq Cancer Hotspot v2 (50 genes, Appendix A), performed in our institution on Ion Torrent/ PGM System (Life Technologies, Camarillo, CA, USA) as previously described (22).Patients enrolled after August 2017 to January 31, 2022, underwent tumour sequencing via FoundationOne CDx (FM1) after an institutional change in panel use.Formalin-fixed, paraffin-embedded (FFPE) tissue was processed per institutional practice, and slides were submitted to Foundation Medicine (Cambridge, MA, USA) for testing.If adequate tissue was not available in patients with no plans for further biopsy, two tubes of whole blood were submitted to Foundation Medicine for analysis under the FoundationOne Liquid CDx platform.Variants identified by FoundationOne CDx are represented as pathogenic or likely pathogenic and variants of unknown significance.For the purposes of this analysis, the variants of uncertain significance were excluded.FoundationOne CDx specimens were also simultaneously profiled for TMB as well as MSI status (23).

Treatment selection
Patient sequencing results were discussed in the NCIS Molecular Tumour Board to identify potential matched therapies or clinical trials.Patients were counselled on outcomes of sequencing and molecular tumour board discussion and enrolled on matched trials, if available.All patients provided informed consent prior to joining a given clinical trial, all of which were approved by the appropriate local Research and Ethics Committee.If matched trials were not available in the unit, patients could undergo matched treatment off trial or unmatched treatment on or off trial after discussion with their treating oncologist.

Definition of matched treatment
Treatment was defined as matched if a molecular alteration detected on sequencing was targeted, or within the pathway targeted, by the administered drug.Treatment was considered non-matched if no molecular alterations were detected or if alterations detected, or the pathways within which they lie, were not targeted by administered treatment.The expression status of ER/progesterone receptor (PR) and CerbB2 was not taken into consideration.Specific scenarios included the following: a. Anti-HER2 therapy is considered matched if ERBB2 alteration or amplification was detected on sequencing analysis regardless of immunohistochemistry results.b.Therapeutics targeting the DNA damage response (DDR) pathway including PARP inhibitors and ATR inhibitors would be considered matched in the presence of mutations in the following list of homologous recombination-related genes previously described by

Follow-up
Patients enrolled on clinical trials underwent follow-up and tumour response assessment per trial protocol, while those treated outside of clinical studies were followed up per standard institution clinical practice.

Evaluable patients
Patients with successful NGS profiling were included in baseline characteristic assessment and mutational profile analysis.Patients who commenced on a new line of systemic therapy after sequencing results were available, with completion of at least one treatment cycle by data cut-off (January 31, 2022), were considered evaluable for clinical outcomes.Patients with less than one completed treatment cycle but with objective evidence of a disease progression event were deemed evaluable.Those lost to follow-up or continuing on treatment without disease progression at data cutoff were censored at the last documented clinic review.

Statistical analysis
Baseline patient characteristics were tabulated and summarised with descriptive statistics.PFS was calculated from the date of first administration of the drug to the date of radiological or clinical progression or death from any cause.Response assessment was per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 criteria.For patients with multiple sequential treatment events after sequencing, details of each individual treatment event (including matched status) were recorded and used to calculate a pooled progression-free survival analysis.The hazard ratio (HR) for risk of progression or death was calculated for this pooled assessment with appropriate frailty adjustment for shared identity.
A per-patient analysis of OS (calculated from the date of treatment start post-sequencing to death or censoring) by matched status was performed.Overall survival of patients receiving matched therapy at any point post-sequencing was compared to that of patients never receiving matched treatment post-sequencing.Clinical benefit rate was calculated as the proportion of patients with complete or partial response or more than 24 weeks of stable disease.Statistical analysis and KM plot generation was performed using STATA version 17.0 (STATA Corp., College Station, TX, USA).Oncoplot and Swimmer plot generated using R and R studio (URL https://www.R-project.org/,Boston, MA; URL http://www.rstudio.com/).

Tumour profiling outcomes
Figure 1 shows an oncoplot of molecular profile results of patients with successful sequencing (n = 158).TP53 was the most commonly mutated gene (47%, 75/158 patients), with a significantly higher frequency of TP53 alterations seen in ER-negative vs. ERpositive tumours (65 vs. 38%, p = 0.001).PIK3CA alterations were seen in 43% (68/158) of patients, with 17 patients having more than one PIK3CA mutation.Of these, the hotspot mutation at PIK3CA H1047R was most common, identified in 17% (27/158) of tumours.
The most commonly altered pathway was PI3K/AKT/PTEN, with alterations in one or more of the pathway member genes (PIK3/AKT/PTEN) seen in 62% (98/158) of patients, with no  Of 109 patients sequenced on the FoundationOne CDx platform, TMB was reported in 96 cases, with a mean TMB of 5.04 mt/Mb (range 0.0-40 mt/Mb).High TMB (≥10 mt/Mb) was seen in 12.5% of patients (12/96).No MSI-high cases were detected.
Germline testing was performed in 26% (41/158) of sequenced patients, with pathogenic or likely pathogenic alterations identified in 8/41 cases.Patients with germline alterations who underwent sequencing with FM1 (5/8) had a corresponding somatic mutation identified in all cases.Three patients with known germline mutations in either BRCA1/2 had tumour profiling performed on the AmpliSeq Cancer Hotspot V2 test panel, which did not include BRCA1/2, and thus, somatic BRCA alterations were not identified.

DDR pathway alterations
Of patients with DDR pathway mutations identified, 12/16 underwent evaluable treatment post-sequencing with a pooled total of 26 treatment events.There was no significant difference in PFS in patients with DDR pathway mutations (median PFS = 16.4 weeks) versus those without (median PFS = 12.4 weeks), HR = 0.71 (95% CI 0.38, 1.32, p = 0.282).Matched treatment was recorded for three patients with DDR pathway mutations.Olaparib was administered in combination with intrathecal methotrexate in a patient with germline (and corresponding somatic) BRCA1 pathogenic variant with a PFS of 59 weeks and as a single agent in a patient with germline (and somatic) PALB2 alteration for 24 weeks prior to progression.Treatment with an ATR inhibitor in the setting of a clinical trial gave a relatively short PFS of 12 weeks in a patient with somatic BRCA1 Q544* (no germline testing performed).One patient received PARPi prior to study enrolment.

TMB and immune checkpoint inhibitors
Immune checkpoint inhibitors (ICIs) were administered alone or in combination to 19 patients (Supplementary Table 2), eight of whom received the combination of anti-PD-L1, AKT inhibitor, and a PARP inhibitor as part of an ongoing clinical trial.These eight patients harboured a mutation in one of PTEN/AKT and/or PIK3CA and were therefore considered matched to the use of an AKT inhibitor.In 5/19 ICI-treated patients, TMB was ≥10, and treatment was considered TMB matched.Median PFS in the 19 ICI-treated patients was 26.0 weeks (95% CI 7, -) with eight patients continuing treatment at the point of data collection (Figure 4).Analysis of this ICI-treated cohort by TMB-matched status showed no significant difference with a median PFS (mPFS) of 44.9 vs. 32.0weeks in TMB-matched vs. unmatched patients, HR = 0.59 (95% CI 0.12, 2.94; p = 0.52).

Discussion
In this retrospective study of a cohort of MBC patients from a tertiary oncology centre in Asia, we report a higher PFS for patients receiving matched versus unmatched treatment, a finding consistent with prior publications (19,24,25).Median PFS in the current study was 21.0 weeks in pooled treatment event analysis, a value that is difficult to compare across publications of early-phase trial cohorts due to the heterogeneous nature of such populations.A prior publication on the IMAC study in 2018, including multiple tumour types, reported a median PFS of 2.9 months, similar to the 3.2 months reported by O'Carrigan and colleagues in a pre-treated MBC cohort (19,22).Our study also yielded a significant improvement in OS in patients receiving matched therapy at any point post-sequencing versus those who do not go on to receive any matched treatment post-sequencing, with HR = 0.45 (95% CI 0.24, 0.84; p = 0.013), highlighting that matching patients based on sequencing outcomes remain beneficial even if the matching cannot occur in the immediate setting.It is therefore vital for specialised drug development units to keep track of patients with targetable molecular mutations and update primary oncologists on new treatment options for matching when available.While OS analysis may be affected by subsequent treatment received and the presence of censored individuals in both matched and unmatched groups, it remains an encouraging result reinforcing the potential benefit of matched therapy in MBC.The frequency of clinical benefit rate (CBR) in the current study was higher for matched treatment events (OR 2.09, p = 0.028), although response rates were similar regardless of matched status at 23% (matched) versus 21% (unmatched) and were overall in keeping with expected responses for pre-treated MBC.
Of all pooled treatment events, 22% were matched based on sequencing results, with the most frequent target being the PIK3/ AKT/PTEN alterations, which were present in 64% of patients.This matching frequency is lower than reported in the pan-cancer IMPACT trial (54%) and in an MBC cohort by O'Carrigan et al. (83%) but similar to that of the Moores Cancer Center PREDICT analysis (20,26).The frequency of matching increased over time, with 11% of treatment events in the first 70 enrolled patients being matched, increasing to 22% for the second 70 patients enrolled and 35% for the most recently recruited group.This may be attributed to the developing nature of the NCIS Developmental Therapeutic Unit over the trial period and a corresponding increase in the number of potential trial options available along with increasing access to approved biomarker-selected treatments such as alpelisib in PIK3CA hotspot mutated tumours.The reported rate in the current study is also contributed by the pooling of all treatment events post-sequencing.If only the first line post-sequencing were examined, the number of patients matched to therapy was 30.5% (32/105) versus 22% if all treatment events were considered.
PIK3, AKT, and PTEN were frequently mutated in the current cohort.Currently approved therapy targeting this pathway includes alpelisib in combination with fulvestrant; however, its use is restricted to patients with HER2-negative MBC harbouring one of 11 PIK3CA hotspot mutations, based on inclusion criteria of the phase III SOLAR-1 trial (16).However, the phase II FAKTION trial points to the benefit of pathway-targeted therapy for patients with a wider span of mutations affecting PIK3/AKT/PTEN when treated with capivasertib in combination with fulvestrant (17).More recently, phase III CAPItello-291 has shown PFS benefit for unselected MBC after progression on aromatase inhibitor with the addition of capivasertib to fulvestrant with the magnitude of benefit more pronounced in AKT pathway-altered patients (18).Results from our current study support the use of matched therapy in a wider pool of PIK3/AKT/PTEN-altered cases with PFS improved in such patients receiving matched treatment compared to those on unmatched therapy (HR = 0.57, p = 0.02).Of patients receiving therapy targeting the PI3K/AKT/PTEN pathway, nine had mutations outside PIK3CA hotspot alterations, with four of nine (44%) having a PFS of ≥12 weeks on matched therapy.Of those matched to therapy, the benefit was enhanced in patients with multiple alterations in PIK3, AKT, or PTEN versus those with single mutations, with HR for progression or death of 0.27 (Figure 3B).This supports previous reports of increased ORR and PFS in patients with multiple PIK3Ca mutations while ongoing treatment with fulvestrant and taselisib (27).This highlights the potential for patients with non-hotspot mutations to benefit from treatment targeting the PI3K/AKT/PTEN pathway and the role that broad-panel NGS has in identifying such alterations.
The number of patients with mutations in the DDR pathway was low (16/158), restricting any analysis of outcomes by treatment status.One patient (IMAC 660, Supplementary Table 1) with a somatic and germline-detected PALB2 alteration had a PFS of 24 weeks with olaparib despite being heavily pre-treated (eight lines prior to therapy).Although anecdotal, this case highlights the expanded role of PARP inhibition outside of germline BRCA1/2altered HER2-negative MBC.The benefit of PARPi outside of current approval limits has been reported in a phase I trial combining olaparib and capivasertib as well as phase II singleagent trials TBCRC-048 and Talazoparib Beyond BRCA (15,28,29).
We report a subgroup of patients who received ICI treatment for MBC (19/105 patients) of whom five had high TMB.The mPFS of 26 weeks in this ICI-treated group is promising when compared to those of trials of enriched subgroups, such as KEYNOTE-158, which reported an mPFS of 4.1 months in non-colorectal MSI-high cancers treated with pembrolizumab, and suggests that there remains a small group of MBC patients who can gain significant benefit from ICI (9).The role of immunotherapy in MBC is currently limited to first-line indication for pembrolizumab in combination with chemotherapy in PD-L1-positive, treatmentnaïve triple-negative breast cancer (TNBC) in combination with chemotherapy, whereas its use in later-line settings as single-agent therapy has not shown significant improvement in treatment outcomes (30,31).In the treatment-refractory setting, pembrolizumab has agnostic approval in the presence of dMMR/ MSI-H or TMB ≥ 10 mt/Mb (8, 9).Numerous ongoing studies continue to evaluate the application of ICI in MBC while attempting to derive an optimal biomarker (32).
Our study has several strengths and limitations.This is one of the first reports of sequencing results and clinical outcomes in an Asianpredominant population and was set within a tertiary referral centre, which enabled access to a range of novel targeted therapies.This is of importance given the number of prospective clinical trials that are being carried out, at least in part, in Asian cancer centres and allows investigators to contextualise the molecular profile of breast cancer in this region.Sequencing was performed on validated commercial platforms ensuring robust reproducible results obtained.
The pooled analysis of treatment events risks incorporating bias where patients receive multiple subsequent treatment lines, each analysed as an individual event.This was adjusted using the frailty adjustment for shared identity in STATA analysis.The number of patients enrolled thus far has not allowed for powering to explore matching at each treatment line, which may be of interest in future studies when larger cohorts of patients with longer-term follow-up are available for analysis.The retrospective nature of the analysis risks selection bias, and low numbers of DDR pathway mutations are in part a reflection of the sample size and restrict any comparative assessment of this subgroup.The use of a more restricted sequencing panel prior to the institutional change to FM1 means that specific alterations will have been unidentified and underreported, including somatic BRCA1/2 mutations, as these genes were not included in the 50 gene panel list of ACH v2 (Supplementary Appendix A).Similarly, TMB and MSI status were not assessed by ACH v2.Mutation detection may have also been affected by tumour heterogeneity, which has been well-described in breast cancer (33,34).The majority of patients in the current studied cohort underwent somatic sequencing on archival tissue.It is possible the mutational profile of a tumour may be significantly different at the point of study entry in comparison to the date of archival tissue acquisition due to the temporal evolution of genomic alterations.Furthermore, the spatial evolution of clones and subclones means that the biopsy site may have affected alterations detected in our cohort.
In conclusion, this study highlights the feasibility of somatic NGS in MBC in facilitating therapeutic matching, primarily in patients with PIK3/AKT/PTEN alterations, which were seen in the majority.Matched therapy is shown to be associated with an improvement in PFS and OS over unmatched treatment in this cohort.

FIGURE 1
FIGURE 1Oncoplot of results for patients with successful sequencing (n = 158).Top 20 most frequently mutated genes shown.A total of 12 patients had no mutations identified.Arranged by oestrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status.
FIGURE 2 (A).Kaplan-Meier pooled analysis of progression-free survival (PFS) by matched versus unmatched treatment (n = 216 treatment events).(B) Kaplan-Meier per-patient analysis of overall survival (OS) by matched versus unmatched status of post-sequencing treatment (n = 105 patients).Survival favours patients receiving matched therapy at any point post-sequencing versus those never receiving matched therapy.HR, hazard ratio.
FIGURE 3 (A) Kaplan-Meier pooled analysis of progression-free survival (PFS) by matched treatment status in PIK3/AKT/PTEN-altered and HER2-negative subgroup.HR, hazard ratio; HER2, human epidermal growth factor receptor 2. (B) Kaplan-Meier pooled analysis of PFS by presence of multiple versus single PIK3/AKT/PTEN alterations for patients with at least one PIK3/AKT/PTEN alteration on matched therapy.HR, hazard ratio.

FIGURE 4
FIGURE 4Swimmer plot of patients treated with immune checkpoint inhibitors (ICIs) arranged by matched status.Matched TMB = 1 represents patients matched to ICI therapy based on the presence of high tumour mutational burden (TMB ≥ 10 mt/Mb).Arrows indicate ongoing treatment at time of data cut-off.