Methodology for Good Machine Learning with Multi-Omics Data

In 2020, Novartis Pharmaceuticals Corporation and the U.S. Food and Drug Administration (FDA) started a 4-year scientific collaboration to approach complex new data modalities and advanced analytics. The scientific question was to find novel radio-genomics-based prognostic and predictive factors for HR+/HER- metastatic breast cancer under a Research Collaboration Agreement. This collaboration has been providing valuable insights to help successfully implement future scientific projects, particularly using artificial intelligence and machine learning. This tutorial aims to provide tangible guidelines for a multi-omics project that includes multidisciplinary expert teams, spanning across different institutions. We cover key ideas, such as "maintaining effective communication" and "following good data science practices," followed by the four steps of exploratory projects, namely (1) plan, (2) design, (3) develop, and (4) disseminate. We break each step into smaller concepts with strategies for implementation and provide illustrations from our collaboration to further give the readers actionable guidance.

Intrinsic Subtype and Overall Survival of Patients with Advanced HR+/HER2- Breast Cancer Treated with Ribociclib and ET: Correlative Analysis of MONALEESA-2, -3, -7

PURPOSE

The MONALEESA-2, -3, -7 trials demonstrated statistically significant and clinically meaningful progression-free survival and overall survival (OS) benefits with ribociclib plus endocrine therapy (ET) versus ET alone in hormone receptor-positive, HER2-negative (HR+/HER2-) advanced breast cancer (ABC). Understanding the association of intrinsic subtypes with survival outcomes could potentially guide treatment decisions. Here, we evaluated the association of intrinsic subtypes with OS in MONALEESA-2, -3, -7.

EXPERIMENTAL DESIGN

Tumor samples from MONALEESA-2, -3, -7 underwent PAM50-based subtyping. The relationship between subtypes and OS was assessed using univariable and multivariable Cox proportional hazards models. Multivariable models were adjusted for clinical prognostic factors.

RESULTS

Overall, 990 tumors (among 2,066 patients) from ribociclib (n = 580) and placebo (n = 410) arms were profiled. Subtype distribution was luminal A, 54.5%; luminal B, 28.0%; HER2-enriched (HER2E) 14.6%; and basal-like, 2.8%; and was consistent across treatment arms. The luminal A subtype had the best OS outcomes in both arms, while basal-like had the worst. Patients with HER2E (HR, 0.60; P = 0.018), luminal B (HR, 0.69; P = 0.023), and luminal A (HR, 0.75; P = 0.021) subtypes derived OS benefit with ribociclib. Patients with basal-like subtype did not derive benefit from ribociclib (HR, 1.92; P = 0.137); however, patient numbers were small (n = 28).

CONCLUSIONS

The prognostic value of intrinsic subtypes for OS was confirmed in this pooled analysis of the MONALEESA trials (largest dataset in HR+/HER2- ABC). While basal-like subtype did not benefit, a consistent OS benefit was observed with ribociclib added to ET across luminal and HER2E subtypes.

Quantitative Assessment of Ribociclib Exposure-Response Relationship to Justify Dose Regimen in Patients with Advanced Breast Cancer

Ribociclib in combination with endocrine therapy (ET) is a globally approved treatment option for patients with hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2-) advanced breast cancer (ABC) and has demonstrated significantly improved overall survival (OS) in 3 phase 3 clinical trials. To justify the dose regimen and dose modification scheme for patients with ABC, the pharmacokinetic (PK), safety, and efficacy data of ribociclib were analyzed. The data of several phase 1-3 clinical studies were pooled and analyzed to characterize the relationship between exposure (dose or PK) and efficacy (progression-free survival (PFS), time to response, and OS) or safety (neutropenia and QT interval prolongation). The exposure-efficacy analysis showed no apparent relationship between ribociclib exposure and efficacy (PFS and OS), and efficacy analysis by dose reduction showed that patients with ABC continued to benefit from the treatment following dose reduction, supporting the starting dose of 600 mg as well as dose reductions to 400 and 200 mg. The exposure-safety analysis showed that neutropenia and QT prolongation are related to ribociclib exposure that can be effectively managed by individualized dose modification (dose reduction/interruption). Collective evidence from the exposure-response analyses for efficacy and safety support the use of ribociclib in combination with ET partners at the starting dose of 600 mg, and also the effectiveness of individualized dose reductions in managing safety, while maintaining efficacy, in patients with HR+/HER2- ABC. This analysis illustrates the utility of quantitative assessment in justifying dose selection and dose modification for oncology medicines.

Updated overall survival from the MONALEESA-3 trial in postmenopausal women with HR+/HER2- advanced breast cancer receiving first-line ribociclib plus fulvestrant

BACKGROUND

The phase III MONALEESA-3 trial included first- (1L) and second-line (2L) patients and demonstrated a significant overall survival (OS) benefit for ribociclib + fulvestrant in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2-) advanced breast cancer (ABC) in the final protocol-specified and exploratory (longer follow-up) OS analyses. At the time of these analyses, the full OS benefit of 1L ribociclib was not completely characterized because the median OS (mOS) was not reached. As CDK4/6 inhibitor (CDK4/6i) + endocrine therapy (ET) is now a preferred option for 1L HR+/HER2- ABC, we report an exploratory analysis (median follow-up, 70.8 months; 14.5 months longer than the prior analysis) to fully elucidate the OS benefit in the MONALEESA-3 1L population.

METHODS

Postmenopausal patients with HR+/HER2- ABC were randomized 2:1 to 1L/2L fulvestrant + ribociclib or placebo. OS in 1L patients (de novo disease or relapse > 12 months from completion of [neo]adjuvant ET) was assessed by Cox proportional hazards model and Kaplan-Meier methods. Progression-free survival 2 (PFS2) and chemotherapy-free survival (CFS) were analyzed. MONALEESA-3 is registered with ClinicalTrials.gov (NCT02422615).

RESULTS

At data cutoff (January 12, 2022; median follow-up time, 70.8 months), mOS was 67.6 versus 51.8 months with 1L ribociclib versus placebo (hazard ratio (HR) 0.67; 95% CI 0.50-0.90); 16.5% and 8.6% of ribociclib and placebo patients, respectively, were still receiving treatment. PFS2 (HR 0.64) and CFS (HR 0.62) favored ribociclib versus placebo. Among those who discontinued treatment, 16.7% and 35.0% on ribociclib or placebo, respectively, received a subsequent CDK4/6i. No new safety signals were observed.

CONCLUSIONS

This analysis of MONALEESA-3 reports the longest mOS thus far (67.6 months) for 1L patients in a phase III ABC trial. These results in a 1L population show that the OS benefit of ribociclib was maintained through extended follow-up, further supporting its use in HR+/HER2- ABC.

Potential value of ctDNA monitoring in metastatic HR + /HER2 - breast cancer: longitudinal ctDNA analysis in the phase Ib MONALEESASIA trial

BACKGROUND

There is increasing interest in the use of liquid biopsies, but data on longitudinal analyses of circulating tumor DNA (ctDNA) remain relatively limited. Here, we report a longitudinal ctDNA analysis of MONALEESASIA, a phase Ib trial evaluating the efficacy and safety of ribociclib plus endocrine therapy (ET) in Asian patients with hormone receptor-positive, human epidermal growth factor receptor-2-negative advanced breast cancer.

METHODS

MONALEESASIA enrolled premenopausal and postmenopausal Japanese and postmenopausal non-Japanese Asian patients. All patients received ribociclib with ET (letrozole, fulvestrant, or tamoxifen with goserelin). ctDNA was analyzed using a targeted next-generation sequencing panel of 572 cancer-related genes and correlated by best overall response (BOR).

RESULTS

Five hundred seventy-four cell-free DNA samples from 87 patients were tested. The most frequently altered genes at baseline included PIK3CA (29%) and TP53 (22%). Treatment with ribociclib plus ET decreased ctDNA in most patients at the first on-treatment time point, regardless of dose or ET partner. Patients with partial response and stable disease had lower ctDNA at baseline that remained low until data cutoff if no progressive disease occurred. Most patients with progressive disease as the best response had higher ctDNA at baseline that remained high at the end of treatment. For patients with partial response and stable disease with subsequent progression, ctDNA increased towards the end of treatment in most patients, with a median lead time of 83 days (14-309 days). In some patients with BOR of partial response who experienced disease progression later, specific gene alterations and total ctDNA fraction increased; this was sometimes observed concurrently with the development of new lesions without a change in target lesion size. Patients with alterations in PIK3CA and TP53 at baseline had shorter median progression-free survival compared with patients with wild-type PIK3CA and TP53, 12.7 and 7.3 months vs 19.2 and 19.4 months, respectively (P = .016 and P = .0001, respectively).

CONCLUSIONS

Higher ctDNA levels and PIK3CA and TP53 alterations detected at baseline were associated with inferior outcomes. On-treatment ctDNA levels were associated with different patterns based on BOR. Longitudinal tracking of ctDNA may be useful for monitoring tumor status and detection of alterations with treatment implications.

TRIAL REGISTRATION

ClinicalTrials.gov NCT02333370 . Registered on January 7, 2015.

Rationale and trial design of NATALEE: a Phase III trial of adjuvant ribociclib + endocrine therapy versus endocrine therapy alone in patients with HR+/HER2- early breast cancer

Background

Ribociclib has demonstrated a statistically significant overall survival benefit in pre- and postmenopausal patients with hormone receptor positive/human epidermal growth factor receptor 2 negative (HR+/HER2-) advanced breast cancer. New Adjuvant Trial with Ribociclib [LEE011] (NATALEE) is a trial evaluating the efficacy and safety of adjuvant ribociclib plus endocrine therapy (ET) versus ET alone in patients with HR+/HER2- early nonmetastatic breast cancer (EBC).

Methods/design

NATALEE is a multicenter, randomized, open-label, Phase III trial in patients with HR+/HER2- EBC. Eligible patients include women, regardless of menopausal status, and men aged ⩾18 years. Select patients with stage IIA, stage IIB, or stage III disease (per the anatomic classification in the AJCC Cancer Staging Manual, 8th edition) with an initial diagnosis ⩽18 months prior to randomization are eligible. Patients receiving standard (neo)adjuvant ET are eligible if treatment was initiated ⩽12 months before randomization. Patients undergo 1:1 randomization to ribociclib 400 mg/day (3 weeks on/1 week off) +ET (letrozole 2.5 mg/day or anastrozole 1 mg/day [investigator's discretion] plus goserelin [men or premenopausal women]) or ET alone. Ribociclib treatment duration is 36 months; ET treatment duration is ⩾60 months. The primary end point is invasive disease-free survival.

Discussion

The 36-month treatment duration of ribociclib in NATALEE is extended compared with that in other adjuvant cyclin-dependent kinases 4 and 6 (CDK4/6) inhibitor trials and is intended to maximize efficacy due to longer duration of CDK4/6 inhibition. Compared with the 600-mg/day dose used in advanced breast cancer, the reduced ribociclib dose used in NATALEE may improve tolerability while maintaining efficacy. NATALEE includes the broadest population of patients with HR+/HER2- EBC of any Phase III trial currently evaluating adjuvant CDK4/6 inhibitor treatment.

Trial registration

ClinicalTrials.gov identifier: NCT03701334 (https://clinicaltrials.gov/ct2/show/NCT03701334).

Abstract P4-01-42: Pooled analysis of post-progression treatments after first-line ribociclib + endocrine therapy in patients with HR+/HER2− advanced breast cancer in the MONALEESA-2, -3, and -7 studies

Background: The MONALEESA (ML) studies showed significant PFS & OS benefits for 1L ribociclib (RIB) + endocrine therapy (ET) in patients (pts) with pre/peri & postmenopausal advanced breast cancer. The benefit of RIB beyond study treatment (tx) was also observed, with improvements in PFS2 & delays in time to 1st subsequent chemotherapy (CT). While there is currently no preferred tx for the next line post-progression on a CDK4/6 inhibitor (CDK4/6i), except alpelisib in pts with a PIK3CA mutation, guidelines encourage multiple lines of ET or ET-based therapies before switching to CT (except for visceral crisis). This pooled exploratory analysis of the ML studies examined outcomes of various tx strategies post progression on RIB + ET.

Methods: Data from pts receiving 1L therapy in ML-2, -3, & -7 (NSAI cohort only & excluding pts with early relapse [≤ 12 mo after end of (neo)adjuvant ET] whose prognosis is closer to that of 2L pts) were pooled & pts receiving 1st subsequent therapies after progression were analyzed. Three groups of subsequent therapies were assessed: ET only, CT, & targeted therapy. Subsequent CT comprises CT +/- any other therapy; targeted therapy includes CDK4/6i, mTORi, PI3Ki, AKTi, etc, +/- ET. Subsequent CT & targeted therapy groups are mutually exclusive. Median duration of study tx, 1st subsequent therapy, & OS (from randomization to death) were analyzed by KM methods. Weighted Cox regressions were performed using inversed propensity scoring matching method (inverse probability tx weighting [IPTW]) to ensure compatible pt characteristics between tx arms. These are not randomized comparisons; only baseline characteristics were used for the estimation of propensity scores in the IPTW, imbalance of prognostic factors at progression may exist.

Results: Median follow-up time was 74 mo. 461 pts treated with RIB (81%) & 440 (86%) with PBO discontinued study tx & received a subsequent therapy. In the RIB arms, the most common 1st subsequent therapies were ET only (40%), CT (29%), combination with targeted therapy (28%), & other (4%); for the PBO arms, 34% received CT as a 1st subsequent therapy & 31% each received ET only or combination with targeted therapy (5% received other). In 14% & 20% of pts in the RIB & PBO arms, the 1st subsequent therapy was a CDK4/6i, of these 31% & 12% were RIB. In general, regardless of type of 1st subsequent therapy, the duration of both the study tx & the 1st subsequent therapy was longer for pts treated with RIB vs PBO (Table). In both RIB & PBO arms, pts who received subsequent CT had the shortest duration on study tx, whereas those who received subsequent targeted therapy combination had the longest. Among pts on 1L RIB + ET, after matching pre-randomization baseline characteristics, subsequent CDK4/6i use was associated with the longest mOS (84 [84-NE] mo), followed by ET only (60 [51-68] mo), then a non-CDK4/6i targeted therapy (52 [43-72] mo); post-progression CT was associated with the shortest mOS (37 [32-48] mo).

Conclusions: This large, pooled analysis of the ML studies shows that, in general, duration of any subsequent therapy was numerically longer post-1L RIB + ET vs PBO + ET, & subsequent CT was used less frequently for pts on RIB vs PBO. Both findings confirm that upfront tx with RIB does not worsen pt outcomes. This trend in enhancement of outcomes of subsequent therapies seen with 1L RIB suggests a post-tx effect that merits further exploration.

Abstract GS2-01: Overall survival subgroup analysis by metastatic site from the phase 3 MONALEESA-2 study of first-line ribociclib + letrozole in postmenopausal patients with advanced HR+/HER2− breast cancer

Background: MONALEESA-2 (ML-2) recently reported a statistically significant overall survival (OS) benefit with first-line ribociclib (RIB) + letrozole (LET) over placebo (PBO) + LET in postmenopausal patients with HR+/HER2- advanced breast cancer (ABC) (median, 63.9 vs 51.4 months; hazard ratio, 0.76; 95% CI, 0.63-0.93; P = .004). Understanding OS outcomes in clinically relevant subgroups of patients is important for improving personalized care and prognosis. Here, we report the results of a prespecified exploratory OS analysis in select patient subgroups by baseline location and number of metastatic sites.. Methods: Postmenopausal patients with HR+/HER2− ABC were randomized 1:1 to receive first-line RIB or PBO with LET. Prespecified exploratory OS analyses were performed for subgroups of special interest by baseline location (bone only [yes or no], liver involvement [yes or no], liver or lung involvement [yes or no]) and number of metastatic sites (< 3 or ≥ 3). The data are hypothesis generating since this analysis was exploratory and not powered for statistical significance.. Results: A total of 668 patients were included in the analysis. A consistent improvement in OS was observed with RIB + LET vs PBO + LET in all subgroups regardless of baseline metastatic site (Table). RIB + LET demonstrated an OS benefit over PBO + LET in clinically relevant subgroups by baseline location and number of metastatic sites, including subgroups of patients with liver metastases, liver or lung metastases, and ≥ 3 metastatic sites, who generally have a worse prognosis.. Conclusion: Consistent with the intent-to-treat population of ML-2, the results of this prespecified exploratory analysis demonstrated an OS benefit with RIB + LET independent of the site and number of metastatic lesions.

TableTreatment Arm (n)HR (95% CI)Bone-only metastasisYesRIB + LET (69)0.78 (0.50-1.21)PBO + LET (79)NoRIB + LET (265)0.77(0.61-0.96)PBO + LET (255)Liver involvementYesRIB + LET (59)0.81 (0.54-1.24)PBO + LET (72)NoRIB + LET (275)0.77 (0.62-0.97)PBO + LET (262)Liver or lung involvementYesRIB + LET (182)0.81(0.62-1.05)PBO + LET (190)NoRIB + LET (152)0.71 (0.53-0.96)PBO + LET (144)No. of metastatic sites< 3RIB + LET (220)0.78(0.61-1.00)PBO + LET (222)≥ 3RIB + LET (114)0.71(0.51-0.98)PBO + LET (112)

Citation Format: Joyce O'Shaughnessy, Salomon M Stemmer, Howard A Burris, Yoon-Sim Yap, Gabe Sonke, Lowell Hart, Mario Campone, Katarina Petrakova, Eric P Winer, Wolfgang Janni, Pierfranco Conte, David A Cameron, Fabrice André, Carlos Arteaga, Juan Pablo Zarate, Arunava Chakravartty, Tetiana Taran, Fabienne Le Gac, Paolo Serra, Gabriel N Hortobagyi. Overall survival subgroup analysis by metastatic site from the phase 3 MONALEESA-2 study of first-line ribociclib + letrozole in postmenopausal patients with advanced HR+/HER2− breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr GS2-01.

Abstract GS2-00: Correlative analysis of overall survival by intrinsic subtype across the MONALEESA-2, -3, and -7 studies of ribociclib + endocrine therapy in patients with HR+/HER2− advanced breast cancer

Background: The MONALEESA (ML)-2, -3, and -7 trials have shown a significant benefit in overall survival (OS) with ribociclib (RIB) + endocrine therapy (ET) over placebo (PBO) + ET in HR+/HER2− advanced breast cancer. HR+ breast cancer is a clinically and biologically heterogeneous disease, with identified intrinsic subtypes that vary in incidence, survival rate, and response to treatment. In a pooled analysis of the ML studies, patients with both luminal and HER2-enriched (HER2E) subtypes exhibited a consistent progression-free survival benefit with RIB + ET. The HER2E subtype (RIB, 14%; PBO, 11%), which is associated with ET resistance and poor outcomes, exhibited the greatest relative reduction in risk of progression or death (61%) with RIB + ET. Here, we report a pooled analysis of the ML-2, -3, and -7 trials, presenting OS by intrinsic subtype.. Methods: Tumor samples from patients enrolled in the ML-2, -3, and -7 trials underwent PAM50-based subtyping (blinded from clinical data), and the correlation between intrinsic subtype and OS was analyzed. Gene expression profiling of formalin-fixed, paraffin-embedded tumor samples was performed using a customized NanoString nCounter GX 800-gene panel. The prognostic and/or predictive relationship between PAM50-based subtypes and OS was evaluated using univariable and multivariable Cox proportional hazards models. Multivariable models were adjusted for known clinical prognostic factors, including age, prior chemotherapy, prior ET, ECOG performance status, visceral disease (presence of liver/lung metastases), bone-only metastases, histological grade, number of metastatic sites, tumor type, and de novo metastatic disease.. Results: From the pooled patient population (N = 2066), 997 tumor (71% primary) samples from the RIB (n = 585) and PBO (n = 412) arms of the ML trials (ML-2, n = 318; ML-3, n = 414; ML-7, n = 265) were profiled. Subtype distribution was consistent across treatment arms (Table). A similar benefit with RIB vs PBO was observed in the ITT (hazard ratio [HR], 0.76; 95% CI, 0.67-0.86) and biomarker (HR, 0.75; 95% CI, 0.63-0.89) populations. In both univariable and multivariable analyses, intrinsic subtype was prognostic for OS in both the RIB and PBO arms (P < .0001 for both arms); patients with luminal A subtype had the best OS outcomes in both arms, whereas patients with basal-like subtype had the worst OS outcomes. Intrinsic subtype was also predictive of OS (subtype-treatment interaction: P = .016 [univariable], P = .007 [multivariable]) with a consistent OS benefit with RIB treatment in all subtypes except for basal-like (Table). Patients with HER2E (HR, 0.60; P = .018), luminal B (HR, 0.69; P = .023), and luminal A (HR, 0.75; P = .021) subtypes all derived benefit from RIB. In patients with the basal-like subtype (n = 30), the HR was 1.89 (P = .148); these results should be interpreted with caution due to the small sample size (3% in each arm) and exploratory nature of this analysis.Conclusions: This pooled analysis of the ML trials confirmed the prognostic and predictive value of intrinsic subtype (PAM50 based) for OS. The addition of RIB to ET resulted in consistent OS benefit across all subtypes except for basal-like. The. consistent survival benefit in the HER2E subtype, which is associated with endocrine resistance and a very poor prognosis compared with luminal disease, warrants further investigation.

SubtypeTreatment ArmDistribution, n (%)OS, median (95% CI), monthsHR (95% CI)P ValueaLuminal ARIB320 (55)68.0 (61.5-NR)0.75 (0.58-0.96).021PBO222 (54)54.6 (48.3-66.2)Luminal BRIB154 (26)58.8 (48.3-79.2)0.69 (0.50-0.95).023PBO124 (30)44.9 (35.5-52.6)HER2-enrichedRIB95 (16)40.3 (33.4-49.0)0.60 (0.40-0.92).018PBO52 (13)29.4 (23.9-42.0)Basal-likeRIB16 (3)19.4 (10.7-33.2)1.89 (0.80-4.47).148PBO14 (3)21.2 (12.8-NR)aP values are all two-sided. NR, not reached.

Citation Format: Lisa Carey, Nadia Solovieff, Fabrice André, Joyce O'Shaughnessy, David A Cameron, Wolfgang Janni, Gabe S Sonke, Yoon-Sim Yap, Denise A Yardley, Juan Pablo Zarate, Tetiana Taran, Faye Su, Agnes Lteif, Aleix Prat. Correlative analysis of overall survival by intrinsic subtype across the MONALEESA-2, -3, and -7 studies of ribociclib + endocrine therapy in patients with HR+/HER2− advanced breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr GS2-00.

Ribociclib plus fulvestrant for postmenopausal women with hormone receptor-positive, human epidermal growth factor receptor 2-negative advanced breast cancer in the phase III randomized MONALEESA-3 trial: updated overall survival

BACKGROUND

Ribociclib plus fulvestrant demonstrated significant progression-free survival (PFS) and overall survival (OS) benefits in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2-) advanced breast cancer (ABC). Here we present a new landmark in survival follow-up for a phase III cyclin-dependent kinases 4 and 6 inhibitor clinical trial in patients with ABC (median, 56.3 months).

PATIENTS AND METHODS

This phase III, randomized, double-blind, placebo-controlled trial was conducted at 174 sites (30 countries). Patients were men and postmenopausal women (age ≥18 years) with histologically/cytologically confirmed HR+/HER2- ABC. Patients could have received ≤1 line of endocrine therapy (ET) but no chemotherapy for ABC. Patients, assigned 2:1, were stratified by the presence/absence of liver/lung metastases and previous ET. Patients received intramuscular fulvestrant (500 mg, day 1 of each 28-day cycle plus day 15 of cycle 1) with oral ribociclib (600 mg/day, 3 weeks on, 1 week off) or placebo. Efficacy analyses were by intention to treat. Safety was assessed in patients receiving ≥1 dose study treatment. OS was a secondary endpoint. MONALEESA-3 is registered with ClinicalTrials.gov (NCT02422615; no longer enrolling).

RESULTS

Between 18 June 2015 and 10 June 2016, 726 patients were randomly assigned (484, ribociclib; 242, placebo). At data cut-off (30 October 2020), median OS (mOS) was 53.7 months (ribociclib) versus 41.5 months (placebo) [hazard ratio (HR), 0.73; 95% confidence interval (CI) 0.59-0.90]. Subgroup analyses were consistent with overall population. In the first-line setting, most patients in the ribociclib arm (∼60%) lived longer than median follow-up; mOS was 51.8 months in the placebo arm (HR, 0.64; 95% CI 0.46-0.88). In the second-line setting, mOS was 39.7 months (ribociclib) versus 33.7 months (placebo) (HR, 0.78; 95% CI 0.59-1.04). No apparent drug-drug interaction between ribociclib and fulvestrant or new safety signals were observed.

CONCLUSIONS

This analysis reported extended OS follow-up in MONALEESA-3. mOS was ∼12 months longer in patients with HR+/HER2- ABC treated with ribociclib plus fulvestrant compared with fulvestrant monotherapy.

Updated overall survival (OS) results from the phase III MONALEESA-3 trial of postmenopausal patients (pts) with HR+/HER2- advanced breast cancer (ABC) treated with fulvestrant (FUL) ± ribociclib (RIB)

1001

Background: The Phase III MONALEESA-3 trial (NCT02422615) previously demonstrated a statistically significant improvement in OS with RIB, a cyclin-dependent kinase 4/6 inhibitor (CDK4/6i), plus FUL compared with placebo (PBO) plus FUL as first-line (1L) or second-line (2L) treatment in postmenopausal pts with HR+/HER2− ABC (median, not reached vs 40.0 mo; hazard ratio [HR], 0.72; 95% CI, 0.57-0.92, P =.00455). This analysis was final per the protocol; following the unblinding of the study, pts still on study treatment in the PBO arm were allowed to cross over to the RIB arm. We report an exploratory analysis of OS after an additional median 16.9 mo of follow-up, allowing for further characterization of long-term survival benefits of RIB. Methods: Postmenopausal pts with HR+/HER2− ABC were randomized 2:1 to receive RIB + FUL or PBO + FUL in 1L and 2L settings. Updated OS was evaluated by Cox proportional hazards model and summarized using Kaplan-Meier methods. Additional postprogression endpoints such as progression-free survival 2 (PFS2), time to chemotherapy (CT), and CT-free survival were also evaluated and summarized. Results: At the data cutoff (Oct 30, 2020), the median follow-up was 56.3 mo (min, 52.7 mo) and 68 (14.0%) and 21 (8.7%) patients were still on treatment in the RIB vs PBO arms, respectively. With this extended follow-up, RIB + FUL continued to demonstrate an OS benefit vs PBO + FUL (median, 53.7 vs 41.5 mo; HR, 0.73; 95% CI, 0.59-0.90). RIB + FUL had prolonged OS vs PBO + FUL in the 1L (median, not reached vs 51.8 mo; HR, 0.64; 95% CI, 0.46-0.88) and 2L subgroups (median, 39.7 vs 33.7 mo; HR, 0.78; 95% CI, 0.59-1.04). Subgroup analyses also showed a consistent OS benefit compared with the intent-to-treat (ITT) population for most subgroups. PFS2, time to CT, and CT-free survival for the ITT population favored RIB + FUL (Table). Among pts who discontinued study treatment, 81.9% and 86.4% received a next-line subsequent antineoplastic therapy, while 14.0% and 30.0% received a CDK4/6i as any subsequent line in the RIB vs PBO arms, respectively. No new safety signals were observed. Conclusions: The previously demonstrated robust and clinically meaningful OS benefit with RIB + FUL compared with PBO + FUL was maintained after almost 5 years of follow-up in postmenopausal pts with HR+/HER2− ABC. The OS benefit of RIB was observed in the 1L and 2L subgroups, which further supports the use of RIB in these populations. The results also demonstrated a significant delay in the use of subsequent CT with RIB vs PBO. Clinical trial information: NCT02422615 .[Table: see text]

Effectiveness of Alpelisib + Fulvestrant Compared with Real-World Standard Treatment Among Patients with HR+, HER2-, PIK3CA-Mutated Breast Cancer

BACKGROUND

The BYLieve trial (NCT03056755) confirmed efficacy and safety of alpelisib with fulvestrant for hormone receptor-positive (HR+), human epidermal growth factor receptor-2-negative (HER2-), PIK3CA-mutated advanced breast cancer (ABC), after cyclin-dependent kinase 4/6 inhibitor (CDK4/6i) with an aromatase inhibitor (AI) as immediate prior therapy. Further analyses were performed to compare efficacy from BYLieve with effectiveness of standard treatment in the real-world setting.

MATERIALS AND METHODS

Patients who progressed on a CDK4/6i plus AI and were treated with alpelisib with fulvestrant in BYLieve were matched with a real-world patient cohort who received standard-of-care from a deidentified clinico-genomics database (CGDB). Primary and secondary endpoints were to compare progression-free survival (PFS), estimated by the Kaplan-Meier method, and the proportion of patients remaining progression-free at 6 months, respectively, between the two cohorts.

RESULTS

A total of 855 patients with PIK3CA-mutant disease who had prior CDK4/6i plus hormone therapy were selected from the CGDB; further matching to 120 patients from BYLieve selected 95 patients without exposure to HER2-targeting agents, clinical study drug, or alpelisib. In unadjusted and postmatching results, primary and secondary endpoints favored treatment with alpelisib with fulvestrant in BYLieve more than standard treatments in the real-world cohort. Postadjustment, median PFS for patients treated with alpelisib in BYLieve was 7.3 versus 3.7 months in the real-world cohort, and 6-month PFS was 54.6% versus 40.1%, respectively.

CONCLUSION

Matched/weighted analysis comparing BYLieve with the real-world setting further supports the clinical benefit of alpelisib with fulvestrant for treatment of HR+, HER2-, PIK3CA-mutant ABC after CDK4/6i treatment.

IMPLICATIONS FOR PRACTICE

Approximately 40% of patients with hormone receptor-positive (HR+), human epidermal growth factor receptor-2-negative (HER2-) advanced breast cancer (ABC) have PIK3CA-mutated tumors, which have been associated with endocrine therapy resistance. Alpelisib, an α-selective phosphatidylinositol-3-kinase inhibitor, demonstrated significantly improved progression-free survival in SOLAR-1 and demonstrated clinical efficacy in BYLieve when combined with fulvestrant. Data are limited in comparing the efficacy of alpelisib combined with fulvestrant with effectiveness of standard therapy after CDK4/6i treatment. Using real-world data, this is the first analysis comparing alpelisib combined with fulvestrant with standard treatments for HR+, HER2-, PIK3CA-mutant ABC in the post-CDK4/6i setting.

Alpelisib plus fulvestrant in PIK3CA-mutated, hormone receptor-positive advanced breast cancer after a CDK4/6 inhibitor (BYLieve): one cohort of a phase 2, multicentre, open-label, non-comparative study

BACKGROUND

Alpelisib, a PI3Kα-selective inhibitor and degrader, plus fulvestrant showed efficacy in hormone receptor-positive, HER2-negative, PIK3CA-mutated advanced breast cancer in SOLAR-1; limited data are available in the post-cyclin-dependent kinase 4/6 inhibitor setting. BYLieve aimed to assess alpelisib plus endocrine therapy in this setting in three cohorts defined by immediate previous treatment; here, we report results from cohort A.

METHODS

This ongoing, phase 2, multicentre, open-label, non-comparative study enrolled patients with hormone receptor-positive, HER2-negative, advanced breast cancer with tumour PIK3CA mutation, following progression on or after previous therapy, including CDK4/6 inhibitors, from 114 study locations (cancer centres, medical centres, university hospitals, and hospitals) in 18 countries worldwide. Participants aged 18 years or older with an Eastern Cooperative Oncology Group performance status of 2 or less, with no more than two previous anticancer treatments and no more than one previous chemotherapy regimen, were enrolled in three cohorts. In cohort A, patients must have had progression on or after a CDK4/6 inhibitor plus an aromatase inhibitor as the immediate previous treatment. Patients received oral alpelisib 300 mg/day (continuously) plus fulvestrant 500 mg intramuscularly on day 1 of each 28-day cycle and on day 15 of cycle 1. The primary endpoint was the proportion of patients alive without disease progression at 6 months per local assessment using Response Evaluation Criteria in Solid Tumors, version 1.1, in patients with a centrally confirmed PIK3CA mutation. This trial is registered with ClinicalTrials.gov, NCT03056755.

FINDINGS

Between Aug 14, 2017, and Dec 17, 2019 (data cutoff), 127 patients with at least 6 months' follow-up were enrolled into cohort A. 121 patients had a centrally confirmed PIK3CA mutation. At data cutoff, median follow-up was 11·7 months (IQR 8·5-15·9). 61 (50·4%; 95% CI 41·2-59·6) of 121 patients were alive without disease progression at 6 months. The most frequent grade 3 or worse adverse events were hyperglycaemia (36 [28%] of 127 patients), rash (12 [9%]), and rash maculopapular (12 [9%]). Serious adverse events occurred in 33 (26%) of 127 patients. No treatment-related deaths were reported.

INTERPRETATION

BYLieve showed activity of alpelisib plus fulvestrant with manageable toxicity in patients with PIK3CA-mutated, hormone receptor-positive, HER2-negative advanced breast cancer, after progression on a CDK4/6 inhibitor plus an aromatase inhibitor.

FUNDING

Novartis Pharmaceuticals.

Real-world incidence, duration, and severity of treatment-emergent (TE) neutropenia among patients (pts) with metastatic breast cancer (MBC) treated with ribociclib (RIB) or palbociclib (PAL)

e13048

Background: Neutropenia is the most common adverse event following administration of CDK4/6 inhibitors RIB and PAL for hormone receptor–positive (HR+) MBC. There are limited comparative real-world data on TE neutropenia in pts receiving these agents. Here we report incidence, duration, and severity data on TE neutropenia in such pts from an electronic health record dataset and administrative claims. Methods: This retrospective study comprised 2 mutually exclusive cohorts of pts with MBC receiving RIB or PAL. Pts were matched 1:1 based on age and year of treatment start. Prior baseline activity of ≥6 mo was required. The MarketScan claims databases was used to evaluate incidence rates of TE neutropenia from Jan 1, 2015, to Dec 31, 2018, in pts receiving RIB or PAL. Rate ratio was calculated using a Poisson model. Data on neutropenia severity and duration were obtained from Optum de-identified Electronic Health Record dataset. Neutropenia severity was defined by neutrophil counts from lab tests (grade 1/2, 1000- < 1500/μL; grade 3, 500- < 1000/μL; grade 4 < 500/μL) within the first 180 days of treatment. Neutropenia duration was estimated using Kaplan-Meier analysis and defined as the time between first abnormal neutrophil result and a lab result demonstrating neutropenia resolution. Results: After 1:1 matching, 152 pts from the MarketScan database were included in both the PAL and RIB cohorts; 168 matched pts were included from the Optum dataset. Neutropenia was reported in 38 pts (25%) in the PAL group and 25 pts (17%) in the RIB group. The rate of neutropenia per person–treatment year was 0.5 (95% CI, 0.4-0.7) in PAL pts vs 0.4 (95% CI, 0.3-0.6) in RIB pts. The rate ratio of neutropenia between treatments (PAL vs RIB) was 1.4 (95% CI, 0.8-2.3), which was not statistically significant, likely due to small sample size. Rates of neutropenia by severity with PAL vs RIB were 32% vs 32% for grade 1/2, 35% vs 26% for grade 3, and 4% vs 4% for grade 4, respectively. The rate ratio for grade 3 or grade 4 neutropenia (PAL vs RIB) was 1.3 (95% CI, 0.9-1.8). Median neutropenia duration was 29 vs 20 days ( P< .01) with PAL vs RIB. Conclusions: Treatment of HR+ MBC with RIB and PAL requires optimal management of TE neutropenia. Real-world data showed that pts with MBC receiving PAL had a numerically higher rate of neutropenia than pts receiving RIB. Rates of grade 3 neutropenia were higher with PAL vs RIB, and duration of neutropenia was longer with PAL vs RIB. Economic burden analyses of neutropenia will be presented.

Alpelisib (ALP) + fulvestrant (FUL) in patients (pts) with PIK3CA-mutated (mut) hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2–) advanced breast cancer (ABC) previously treated with cyclin-dependent kinase 4/6 inhibitor (CDKi) + aromatase inhibitor (AI): BYLieve study results

1006

Background: PIK3CA mutations (mut) occur in ~40% of pts with HR+, HER2– ABC and are associated with poor prognosis and resistance to treatment (tx). ALP, a PI3Kα inhibitor, plus FUL demonstrated efficacy in the phase 3 SOLAR-1 trial of HR+, HER2– PIK3CA-mut ABC. Little clinical data and few prospective studies are available to inform tx decisions for pts with HR+, HER2– PIK3CA-mut ABC in the post-CDKi setting. BYLieve is the first trial evaluating ALP + endocrine therapy (ET; FUL or letrozole) in pts with HR+, HER2– PIK3CA-mut ABC who progressed on/after prior therapy, including CDKi. In this ongoing phase 2, open-label, noncomparative study, 112 pts with centrally confirmed PIK3CA mut in tumor tissue are enrolled in each cohort per immediate prior tx of CDKi + AI, CDKi + FUL, or systemic chemo or ET. Enrollment is complete in prior CDKi + AI and CDKi + FUL cohorts and ongoing in prior systemic chemo or ET cohort. We report on the cohort of pts with CDKi + AI as immediate prior tx. Methods: Pts received ALP 300 mg/day + FUL 500 mg Q28D + C1D15. Primary endpoint was proportion of pts alive without disease progression at 6 mo per local assessment; 2-sided 95% confidence intervals (CI) were calculated using Clopper and Pearson (1934) exact method. Evidence of clinically meaningful tx effect was defined as the lower bound of the 95% CI > 30%. Safety was assessed in all patients; AEs presented by preferred term. Results: 127 pts whose immediate prior tx was CDKi + AI were enrolled, of whom 121 had centrally confirmed PIK3CA mut; median follow-up was 11.7 mo. Primary endpoint was met: proportion of pts without disease progression at 6 mo was 50.4% (95% CI, 41.2-59.6). Most frequent all-grade AEs were diarrhea (60%), hyperglycemia (58%), nausea (46%), fatigue (29%), decreased appetite (28%), and rash (28%). Most frequent grade ≥3 AEs included hyperglycemia (28%), rash (9%), and rash maculopapular (9%). Incidence of AEs leading to discontinuation was low; most frequent AEs leading to discontinuation were rash (5 pts, 3.9%), colitis, hyperglycemia, urticaria, and vomiting (2 pts, 1.6% each). Conclusions: With follow-up still ongoing, BYLieve shows in a large number of pts that ALP + FUL demonstrates clinically meaningful efficacy and manageable toxicity post CDKi tx. Building on findings from SOLAR-1, BYLieve further supports use of ALP + FUL for HR+, HER2– PIK3CA-mut ABC. Clinical trial information: NCT03056755 .

The effects of lapatinib on cardiac repolarization: results from a placebo controlled, single sequence, crossover study in patients with advanced solid tumors

PURPOSE

To evaluate the effect of lapatinib on the QTc interval and ECG parameters in patients with advanced solid tumors.

METHODS

This was a multicenter, placebo-controlled study in subjects with advanced solid tumors. Subjects were administered two doses of matching placebo on day 1, 12 h apart and one dose in the morning on day 2. Two doses of lapatinib 2000 mg were administered orally on day 3, 12 h apart and one dose in the morning on day 4. Twelve-lead digital ECGs were extracted from continuous Holter recordings at pre-specified time points over the 24-h period on days 2 and 4. Venous blood samples for lapatinib concentrations were obtained immediately following the ECGs.

RESULTS

A maximum mean baseline-adjusted, placebo time-matched increase in QTcF, (ddQTcF) in the evaluable, (EV) population (n = 37) of 8.8 ms (90% CI 4.1, 13.4) occurred approximately 10 h after the third lapatinib dose. These results were consistent with those in the pharmacodynamic, PD population, (n = 52) (ddQTcF = 7.9 ms; 90% CI 4.1, 11.7). No subject experienced QTcF increases from baseline of > 60 ms on lapatinib or placebo. The geometric mean lapatinib C_max of 3902 ng/mL was observed at 3.6 h post-dose.

CONCLUSIONS

These data show a relevant, treatment-related increase in QTcF after treatment with three doses of lapatinib 2000 mg. This study confirms the need for caution in patients with solid tumors treated with lapatinib, and who are concomitantly receiving drugs that are strong CYP3A inhibitors and/or prolong the QTc.

Leeomic: A comprehensive proteomic analysis towards discovery of predictive patterns of protein expression to ribociclib sensitivity and resistance—A compLEEment-1 Canadian correlative sub-study

TPS3170

Background: Despite developments in the treatment of advanced hormone receptor positive (HR+), human epidermal growth factor receptor 2 negative (HER2-) breast cancer, primary or acquired resistance eventually occurs in all cases and there is still very limited understanding of the mechanisms of resistance to therapy. LEEOMIC is a sub-study of the main CompLEEment-1 ( N = 3255 patients enrolled, CLEE011A2404 v03) trial, an open-label, phase 3b study evaluating ribociclib + letrozole as first-line therapy in an expanded advanced breast cancer patient population which recruited over 250 Canadian patients. The purpose of this Canadian correlative sample collection study is to explore the mechanisms of response and resistance to ribociclib in combination with letrozole through proteomic and ctDNA analysis. Methods: The British Columbia Cancer Research Centre team developed a novel and optimized MS/MS platform called SP3-Clinical Tissue Proteomics (SP3-CTP) to perform in-depth proteome profiling ( > 8,000 proteins) from formalin fixed paraffin embedded (FFPE) material (10-micron section). SP3-CTP analysis of the proteome of the study patients who did not achieve clinical benefit (primary resistance: progression within 3 months of treatment) will be compared to the proteome of the sub-group of prolonged responders (time to progression of 22 months or more) in order to identify biomarkers that can predict response or de-novo resistance to therapy. Archival tumor biopsies (primary or metastatic) collected from the study will be submitted for proteomic analysis to identify proteomic expression levels that may serve as predictor of response. It is anticipated that over 150 samples will be collected. If available, blood samples taken at time of progression or end of treatment will also be analyzed for ctDNA for genetic profiling and to study if there is any correlation between genetic mutations and response or resistance to therapy. Currently, both tissue and blood samples are being collected and no analysis has been conducted thus far. Clinical trial information: NCT03613220.

Assessment of liver and cardiac iron overload using MRI in patients with chronic anemias in Latin American countries: results from ASIMILA study

OBJECTIVES

A multicenter, noninterventional, observational study was conducted in the Latin American countries including Argentina, Brazil, Colombia, Mexico, and Venezuela to assess the prevalence of liver and cardiac iron overload using magnetic resonance imaging (MRI) in patients with chronic anemias except thalassemia.

METHODS

Patients aged >10 years with transfusion-dependent anemias, except thalassemia, either with <20 units of red blood cell (RBC) transfusions with serum ferritin (SF) levels >2000 ng/mL or with ≥20 units of RBC transfusions regardless of SF level in their lifetime, were enrolled. Iron overload was assessed using MRI.

RESULTS

Among 175 patients included, the majority had sickle cell disease (SCD; 52%), followed by aplastic anemia (AA; 17.7%), myelodysplastic syndrome (MDS; 8.6%), Diamond-Blackfan anemia (DBA; 4%), pure red cell aplasia (1.1%), and others (16.6%). Liver iron overload was observed in 76.4% of patients, while cardiac iron overload was seen in 19.2% when assessed by MRI. The prevalence of iron overload was 80.2% in patients with SCD, 73.3% in MDS, 77.4% in AA, 100% in pure red cell aplasia, 71.4% in DBA, and 68.9% in other transfusion-related disorders. A moderate correlation between liver iron concentration (LIC) and SF was observed in patients with SCD and MDS (r = 0.47 and r = 0.61, respectively). All adverse events reported were consistent with the published data for deferasirox or underlying disease.

CONCLUSION

A high prevalence of iron overload in this patient population in Latin American countries indicates that a better diagnosis and management of iron overload is required in these countries.