Purpose:

The PI3K pathway is dysregulated in the majority of triple-negative breast cancers (TNBC), yet single-agent inhibition of PI3K has been ineffective in TNBC. PI3K inhibition leads to an immediate compensatory upregulation of the Wnt pathway. Dual targeting of both pathways is highly synergistic against TNBC models in vitro and in vivo. We initiated a phase I clinical trial combining gedatolisib, a pan-class I isoform PI3K/mTOR inhibitor, and cofetuzumab pelidotin, an antibody–drug conjugate against the cell-surface PTK7 protein (Wnt pathway coreceptor) with an auristatin payload.

Patients and Methods:

Participants (pt) had metastatic TNBC or estrogen receptor (ER) low (ER and PgR < 5%, HER2-negative) breast cancer, and had received at least one prior chemotherapy for advanced disease. The primary objective was safety. Secondary endpoints included overall response rate (ORR), clinical benefit at 18 weeks (CB18), progression-free survival (PFS), and correlative analyses.

Results:

A total of 18 pts were enrolled in three dose cohorts: gedatolisib 110 mg weekly + cofetuzumab pelidotin 1.4 mg/kg every 3 weeks (n = 4), 180 mg + 1.4 mg/kg (n = 3), and 180 mg + 2.8 mg/kg (n = 11). Nausea, anorexia, fatigue, and mucositis were common but rarely reached ≥grade 3 severity. Myelosuppression was uncommon. ORR was 16.7% (3/18). An additional 3 pts had stable disease (of these 2 had stable disease for >18 weeks); CB18 was 27.8%. Median PFS was 2.0 months (95% confidence interval for PFS: 1.2–6.2). Pts with clinical benefit were enriched with genomic alterations in the PI3K and PTK7 pathways.

Conclusions:

The combination of gedatolisib + cofetuzumab pelidotin was well tolerated and demonstrated promising clinical activity. Further investigation of this drug combination in metastatic TNBC is warranted.

Translational Relevance

Targeted therapy options for metastatic triple-negative breast cancer (TNBC) are limited. Previous genomic analyses of large cohorts of TNBC have consistently demonstrated activation of the PI3K pathway. However, single-agent PI3K inhibition has had modest clinical efficacy. Preclinical data from our group and others have demonstrated that upregulation of the Wnt pathway induces resistance to PI3K inhibition. Herein, we report a phase I clinical trial of gedatolisib (PI3K/mTOR inhibitor) in combination with cofetuzumab pelidotin, an antibody–drug conjugate targeting the Wnt pathway receptor PTK7, in patients with metastatic TNBC. In this first clinical trial, combining a PI3K and Wnt pathway agent we report a favorable safety profile and antitumor activity. Further clinical trials testing combinations targeting these two pathways are warranted.

Triple-negative breast cancer (TNBC) is a heterogeneous disease that comprises a minority (15%–20%) of breast cancer cases, but has a disproptionally higher mortality (1). TNBC is characterized by the absence of estrogen-receptor (ER), progesterone-receptor, and HER2 overexpression (2–4). Patients with TNBC have a higher probability of relapse in the first 3 years after surgery, higher rates of visceral metastases, and shortened overall survival (OS) upon onset of metastatic disease when compared with hormone receptor–positive and HER2-positive breast cancers (5). Despite the recent FDA approval of immune checkpoint inhibition using pembrolizumab for early-stage (6) and late-stage disease (7), cytotoxic chemotherapy remains the mainstay treatment for metastatic TNBC. Previous attempts to introduce targeted therapies in TNBC by inhibiting the EGFR and c-KIT receptors were unsuccessful (8–11). PARP inhibitors are effective, but only in the subset of patients with deleterious BRCA1/2 mutations (12, 13). Sacituzumab govitecan, an antibody–drug conjugate (ADC) targeting TROP2 was recently approved by the FDA for metastatic TNBC, based on an overall response rate (ORR) of 33.3% in a single-arm phase II study unselected for TROP2 expression (14). These results were recently confirmed in the phase III ASCENT study demonstrating an ORR of 35% with sacituzumab compared with 5% with single-agent chemotherapy in patients treated with two prior lines of therapy (15). Clearly, there remains a critical need to identify additional novel targeted therapies for TNBC.

The majority of TNBCs (∼70%) harbor a genomic aberration in the PI3K pathway (16), most commonly an activating mutation or somatic copy-number change in one of the canonical components (17). Somewhat surprisingly, single-agent inhibition of PI3K has been ineffective as a treatment for TNBC (18). We have previously reported that PI3K inhibition in TNBC results in an immediate compensatory upregulation of the Wnt pathway (19). Activation of the Wnt pathway is well known for its role in cancer metastases and confers resistance to PI3K inhibition in TNBC, pancreatic, and colorectal cell lines (20–22). Simultaneous dual targeting of both pathways is synergistic against TNBC models in vitro and in vivo (19).

Building on our preclinical observations, we initiated a phase I clinical trial of gedatolisib (23, 24), a pan-class I isoform PI3K/mTOR inhibitor, and cofetuzumab pelidotin (25, 26), an ADC against the cell-surface PTK7 protein with an auristatin payload. PTK7 is a pseudoreceptor tyrosine kinase involved in noncanonical Wnt signaling that maintains developmental planar cell polarity in normal cells (27). PTK7 is overexpressed in a variety of cancers including TNBC (25), and its expression is induced by PI3K inhibition therefore providing rationale for exploring the combination of a PI3K inhibitor with a PTK7-targeting ADC in TNBC. The payload for cofetuzumab pelidotin, auristatin, is in itself synergistic with gedatolisib, providing another potential mechanism of synergy for the combination (28).

Study population

Participants (pt) had metastatic TNBC or ER-low (ER and PgR <5%, HER2-negative) breast cancer, and had received at least one prior chemotherapy for metastatic disease. Pts were excluded if previously treated with a PI3K or an mTOR inhibitor or if they had previous exposure to cofetuzumab pelidotin. Pts were required to have an Eastern Cooperative Oncology Group performance status ≤1 with adequate hematologic, hepatic, and renal function. Pts with treated and stable central nervous system involvement were allowed. Given the potential for hypergylcemia with gedatolisib, pts with uncontrolled diabetes were excluded.

Trial design and statistical analyses

This single-center, phase I dose-escalation trial utilized a traditional 3+3 schema with a small expansion cohort at the recommended phase II dose to better characterize safety (Supplementary Fig. S1). The primary objective was safety as assessed by NCI CTC v4.0 criteria. Dose-limiting toxicity (DLT) was defined as grade 4 neutropenia lasting ≥ 7 days; febrile neutropenia; grade 3 thrombocytopenia associated with bleeding or grade 4 thrombocytopenia lasting ≥ 4 days; any nonhematologic grade 3 ≥ toxicity despite the use of adequate/maximal medical interventions and/or prophylaxis; alanine aminotransferase or aspartate aminotransferase > 3× upper limit of normal (ULN) and total bilirubin >2× ULN with serum alkaline phosphatase normal; any toxicity that resulted in a >14-day delay in treatment; or any death not clearly due to underlying disease or extraneous causes. Secondary endpoints included ORR, clinical benefit at 18 weeks [CB18; prospectively defined as all pts with a complete response, partial response, or stable disease (SD) for at least 18 weeks], and progression-free survival (PFS). Exploratory analyses probed for association between genomic/pathologic features and clinical efficacy to identify putative biomarkers for consideration in subsequent trials. The Kaplan–Meier method was used to analyze for PFS and OS [with median times and 95% confidence intervals (CI) calculated] using SAS Version 9.4. The study was approved by the Institutional Review Board at Indiana University (Indianapolis, IN); all pts provided written informed consent prior to study entry. This study was conducted in accordance with U.S. Common Rule.

Treatment plan

The recommended phase II dose (RP2D) and toxicity profile had been established for each agent as monotherapy. As overlapping toxicity and pharmacokinetic interactions were not expected, we planned a limited dose escalation, starting with 50% of the monotherapy RP2D for each agent in cohort 1. The full RP2D of each agent as monotherapy was delivered in cohort 3 (Table 1). Gedatolisib was infused first followed by cofetuzumab pelidotin. Pts were treated prophylactically with a steroid mouthwash prior to administration of gedatolisib to minimize mucositis. Pts were evaluated clinically each week during the first two cycles and on day 1 of each subsequent cycle. Serum chemistry was obtained at the start of each cycle; complete blood counts were obtained prior to each gedatolisib infusion. Dose modifications for hematologic and nonhematologic toxicity were prespecified. Response was evaluated according to RECIST 1.1 every two cycles (6 weeks) through week 18, then every three cycles thereafter. Tumor biopsies for correlative analyses were obtained at screening and cycle 1 day 15 in pts with accessible lesions.

Table 1.

Treatment plan.

Treatment plan.
Treatment plan.

PTK7 and phospho-AKT IHC

PTK7 IHC was performed by Flagship Biosciences using a proprietary antibody for staining under a protocol developed by Pfizer (29). Staining for phospho-AKT (pAKT, Ser473), was performed using antibody #4060 procured from Cell Signaling Technology as described previously (30). PTK7 and pAKT IHC results were assessed by a certified pathologist and quantified by H-score.

Exome library preparation and sequencing

DNA from tumor and matched blood normal was extracted using the Qiagen AllPrep DNA/RNA FFPE kit and QIAamp DNA Blood Mini kit, respectively. DNA library preparation utilized the SureSelect XTHS Target Enrichment System for Illumina Paired-End Multiplexed Sequencing Version A1, July 2017 (Agilent Technologies). Libraries were then hybridized, captured, and amplified with the Agilent Human All Exon V7 probe set (48 Mb, hg38). Libraries were sequenced on a NovaSeq 6000 sequencer using 150 bp paired-end chemistry (Illumina, Inc.). A Phred quality score (Q score) was used to measure the quality of sequencing. More than 90% of the sequencing reads reached Q30 (99.9% base call accuracy).

Exome variant and copy-number analysis

FASTQ files were aligned to the human reference genome (b37/hg19) using the Burrows-Wheeler Aligner (v. 0.7.17, http://bio-bwa.sourceforge.net/; ref. 31). PCR duplicates were removed and coverage metrics were calculated using Picard tools (v.2.21.2, http://picard.sourceforge.net/). The Genome Analysis Toolkit (GATK, v. 4.1.4.1, http://www.broadinstitute.org/gatk/) was used for SNP and INDEL discovery according to GATK best practices (32). Copy-number variants were analyzed using CODEX2 (v.1.3.0; ref. 33).

Data availability

Raw sequencing data for this study were generated by The Center for Medical Genomics at Indiana University School of Medicine (Indianapolis, IN). Derived data supporting the findings of this study are available in the Supplementary Data.

Participant characteristics

We enrolled 18 pts: 10 in three dose-escalation cohorts and 8 in expansion at cohort 3. Baseline characteristics are shown in Table 2. Pts were heavily pretreated with the majority having prior exposure to an anthracycline, taxane, platinum, and capecitabine.

Table 2.

Participant characteristics.

Overall
Characteristic(N = 18)
Age (year) 
 Median (range) 53 (32–77) 
Gender, n (%) 
 Female 18 (100%) 
Race, n (%) 
 White 17 (94.4%) 
 Black or African American 1 (5.6%) 
Ethnicity 
 Not Latino or Hispanic 17 (94.4%) 
 Unknown 1 (5.6%) 
Prior therapies 
 Taxane 18 (100%) 
 Platinum 18 (100%) 
 Anthracycline 17 (94.4%) 
 Capecitabine 13 (72.2%) 
 Gemcitabine 8 (44.4%) 
 Eribulin 6 (33.3%) 
 Vinorelbine 5 (27.8%) 
 Immunotherapy 4 (22.2%) 
 PARP inhibitor 3 (16.7%) 
 Other 3 (16.7%) 
Overall
Characteristic(N = 18)
Age (year) 
 Median (range) 53 (32–77) 
Gender, n (%) 
 Female 18 (100%) 
Race, n (%) 
 White 17 (94.4%) 
 Black or African American 1 (5.6%) 
Ethnicity 
 Not Latino or Hispanic 17 (94.4%) 
 Unknown 1 (5.6%) 
Prior therapies 
 Taxane 18 (100%) 
 Platinum 18 (100%) 
 Anthracycline 17 (94.4%) 
 Capecitabine 13 (72.2%) 
 Gemcitabine 8 (44.4%) 
 Eribulin 6 (33.3%) 
 Vinorelbine 5 (27.8%) 
 Immunotherapy 4 (22.2%) 
 PARP inhibitor 3 (16.7%) 
 Other 3 (16.7%) 

Safety

Dose escalation was completed with no DLTs; an additional pt was included in cohort 1 to accommodate an immediate clinical need for treatment. A small expansion cohort proceeded at the full RP2D for monotherapy for both agents. The most common adverse events of any grade were nausea (n = 16, 89%), anorexia (n = 13, 72%), constipation (n = 12, 67%), fatigue (n = 12), and mucositis (n = 12). Grade 3 or greater toxicity (nausea, n = 1; fatigue, n = 2) and myelosuppression (grade ≥3 neutropenia, n = 2) were uncommon (Table 3).

Table 3.

Observed toxicities in >10% of pts or observed as grade ≥ 3 using CTCAE.

CTCAE termGrade 1Grade 2Grade 3Grade 4Grade 5TotalTotal percentGrade 3–5Grade 3–5 percent
Nausea 11 16 88.89 5.56 
Anorexia 13 72.22 0.00 
Constipation 10 12 66.67 0.00 
Fatigue 12 66.67 11.11 
Mucositis oral 10 12 66.67 0.00 
Back pain 11 61.11 11.11 
Dyspnea 11 61.11 22.22 
Vomiting 11 61.11 16.67 
Alopecia 10 55.56 0.00 
Pruritus 50.00 0.00 
Diarrhea 44.44 5.56 
Dyspepsia 44.44 0.00 
Headache 44.44 0.00 
Anemia 38.89 11.11 
Cough 38.89 0.00 
Insomnia 38.89 0.00 
Neutrophil count decreased 33.33 11.11 
Anxiety 27.78 0.00 
Hyperglycemia 27.78 0.00 
Pain 27.78 11.11 
Blurred vision 22.22 0.00 
Dysgeusia 22.22 0.00 
Edema limbs 22.22 0.00 
Hypokalemia 22.22 5.56 
Pain in extremity 22.22 0.00 
Peripheral sensory neuropathy 22.22 0.00 
Aspartate aminotransferase increased 16.67 0.00 
Chest wall pain 16.67 0.00 
Chills 16.67 0.00 
Lung infection 16.67 11.11 
Mucosal infection 16.67 5.56 
Rash acneiform 16.67 0.00 
Rash maculopapular 16.67 5.56 
Urinary tract pain 16.67 0.00 
Abdominal pain 11.11 5.56 
Alanine aminotransferase increased 11.11 0.00 
Allergic rhinitis 11.11 0.00 
Bone pain 11.11 0.00 
Dry mouth 11.11 0.00 
Infections and infestations—Other, specify 11.11 0.00 
Lymphedema 11.11 0.00 
Nasal congestion 11.11 0.00 
Oral pain 11.11 0.00 
Paresthesia 11.11 0.00 
Skin and subcutaneous tissue disorders—Other, specify 11.11 0.00 
Thromboembolic event 11.11 0.00 
Urinary tract infection 11.11 0.00 
Urinary urgency 11.11 0.00 
Colitis 5.56 5.56 
Dehydration 5.56 5.56 
Investigations—Other, specify 5.56 5.56 
Respiratory failure 5.56 5.56 
Skin infection 5.56 5.56 
White blood cell decreased 5.56 5.56 
CTCAE termGrade 1Grade 2Grade 3Grade 4Grade 5TotalTotal percentGrade 3–5Grade 3–5 percent
Nausea 11 16 88.89 5.56 
Anorexia 13 72.22 0.00 
Constipation 10 12 66.67 0.00 
Fatigue 12 66.67 11.11 
Mucositis oral 10 12 66.67 0.00 
Back pain 11 61.11 11.11 
Dyspnea 11 61.11 22.22 
Vomiting 11 61.11 16.67 
Alopecia 10 55.56 0.00 
Pruritus 50.00 0.00 
Diarrhea 44.44 5.56 
Dyspepsia 44.44 0.00 
Headache 44.44 0.00 
Anemia 38.89 11.11 
Cough 38.89 0.00 
Insomnia 38.89 0.00 
Neutrophil count decreased 33.33 11.11 
Anxiety 27.78 0.00 
Hyperglycemia 27.78 0.00 
Pain 27.78 11.11 
Blurred vision 22.22 0.00 
Dysgeusia 22.22 0.00 
Edema limbs 22.22 0.00 
Hypokalemia 22.22 5.56 
Pain in extremity 22.22 0.00 
Peripheral sensory neuropathy 22.22 0.00 
Aspartate aminotransferase increased 16.67 0.00 
Chest wall pain 16.67 0.00 
Chills 16.67 0.00 
Lung infection 16.67 11.11 
Mucosal infection 16.67 5.56 
Rash acneiform 16.67 0.00 
Rash maculopapular 16.67 5.56 
Urinary tract pain 16.67 0.00 
Abdominal pain 11.11 5.56 
Alanine aminotransferase increased 11.11 0.00 
Allergic rhinitis 11.11 0.00 
Bone pain 11.11 0.00 
Dry mouth 11.11 0.00 
Infections and infestations—Other, specify 11.11 0.00 
Lymphedema 11.11 0.00 
Nasal congestion 11.11 0.00 
Oral pain 11.11 0.00 
Paresthesia 11.11 0.00 
Skin and subcutaneous tissue disorders—Other, specify 11.11 0.00 
Thromboembolic event 11.11 0.00 
Urinary tract infection 11.11 0.00 
Urinary urgency 11.11 0.00 
Colitis 5.56 5.56 
Dehydration 5.56 5.56 
Investigations—Other, specify 5.56 5.56 
Respiratory failure 5.56 5.56 
Skin infection 5.56 5.56 
White blood cell decreased 5.56 5.56 

Note: There were no grade 5 events observed.

Efficacy

A total of 16 of 18 pts enrolled were evaluable for response; 2 pts discontinued treatment prior to the first disease assessment that was not due to toxicity or progression. Three pts achieved a confirmed partial response (ORR, 16.7%; 3/18 including the two nonevaluable pts). An additional 3 pts had SD; of these 2 had SD for > 18 weeks. The CB18 was 27.8% (5/18; Fig. 1). Of note, 1 pt with SD with Inflammatory TNBC (Pt #: 0613-25) had their only target lesion biopsied and therefore was not eligible for measurement per RECIST; however, they had a prolonged exceptional response of approximately 2 years. The pt has been off all anti-cancer treatment for > 15 months. Overall, after a median follow-up of 7.9 months, median PFS was 2.0 months (95% CI for PFS: 1.2–6.2; Fig. 2A); median OS was 9.4 months (95% CI for OS: 4.3–15.8; Fig. 2B).

Figure 1.

Waterfall plot demonstrating best tumor response from baseline (y-axis). A total of 18 pts enrolled on trial: 16 were evaluable for response; 13 pts with measurable disease reached at least the first prespecified imaging timepoint for disease assessment. Pts are color coded by dose cohort. PD, progressive disease; PR, partial response; SD, stable disease. Of note, the pt with best percent change for the target lesions, unfortunately, in the same visit had increasing/progressive disease for the nontarget lesions, resulting in overall best response as PD.

Figure 1.

Waterfall plot demonstrating best tumor response from baseline (y-axis). A total of 18 pts enrolled on trial: 16 were evaluable for response; 13 pts with measurable disease reached at least the first prespecified imaging timepoint for disease assessment. Pts are color coded by dose cohort. PD, progressive disease; PR, partial response; SD, stable disease. Of note, the pt with best percent change for the target lesions, unfortunately, in the same visit had increasing/progressive disease for the nontarget lesions, resulting in overall best response as PD.

Close modal
Figure 2.

A, Kaplan–Meier plot of PFS demonstrating a median PFS of 2.0 months (95% CI for PFS: 1.2–6.2). B, Kaplan–Meier plot of OS demonstrating a median OS of 9.4 months (95% CI for OS: 4.3–15.8).

Figure 2.

A, Kaplan–Meier plot of PFS demonstrating a median PFS of 2.0 months (95% CI for PFS: 1.2–6.2). B, Kaplan–Meier plot of OS demonstrating a median OS of 9.4 months (95% CI for OS: 4.3–15.8).

Close modal

Correlative analyses

Using whole-exome sequencing, we set out to observe if somatic DNA variants were associated with clinical response (Table 4). Of the 18 pts on trial, 15 had successful whole-exome sequencing of the baseline biopsy sample. We did not detect abberations in the PI3K pathway or PTK7 in 6 of 15 pts. In pts who had a partial response (n = 3), we observed a PTEN I101N mutation in 1 pt and a high gain of AKT2 in another. In pts who had SD (n = 3), one harbored a high gain of AKT2, another had a PIK3CA E545A gain-of-function mutation and PTEN single-copy deletion, and a third had high gains in AKT2, PIK3CG, and PTK7. In those pts with progressive disease (n = 10), we observed an AKT2 S268 L gain-of-function mutation in 1 pt, a PIK3CA H1047R gain-of-function mutation in 1 pt, a PTK7 variant (of unknown significance), and high gains of AKT2 in 3 pts and PIK3CB in 2 pts. Overall, 5 of 5 pts with either a partial response (PR) or SD and matched exome sequencing harbored an aberration in the PI3K pathway or PTK7, compared with 4 of 9 pts with progressive disease.

Table 4.

Genomic alterations and immunohistochemistry results with best response.

Patient IDDose cohortBest responseCB18PFS (months)DNA variantsCopy-number variationPTK7 H-score (screen)PTK7 H-score (C1D15)Nuc-pAKT (screen)Cyto-pAKT (screen)Nuc-pAKT (C1D15)Cyto-pAKT (C1D15)
Progressive disease  1.12 N/A N/A 120 166 10 
Progressive disease  1.22 N/A N/A 16 25 
Progressive disease  1.15 N/A High gain: AKT2, PIK3CB N/A 232 N/A N/A 
Progressive disease  2.1 PTK7 E139Q High gain: AKT2, PIK3CB 101 22 
Progressive disease  0.69 AKT2 S268L N/A 197 N/A 20 N/A N/A 
Partial response Yes 11.31 PTEN I101N N/A 116 N/A 200 10 
Progressive disease  1.22 N/A N/A 224 116 10 
10 Progressive disease  1.84 N/A N/A 68 82 
12 Progressive disease  1.51 N/A N/A 111 N/A 60 N/A N/A 
15 Partial response Yes 6.21 N/A High gain: AKT2 144 N/A 20 N/A N/A 
16 Progressive disease  1.45 PIK3CA H1047R High gain: AKT2 168 N/A 140 N/A N/A 
18 Not evaluable  — N/A N/A N/A N/A N/A N/A N/A N/A 
19 Partial response Yes 7.36 NS NS N/A N/A N/A N/A N/A N/A 
20 Stable disease Yes 7.07 N/A High gain: AKT2 120 120 200 N/A N/A 
21 Progressive disease  3.02 NS NS N/A N/A N/A N/A N/A N/A 
22 Not evaluable  — NS NS N/A N/A N/A N/A N/A N/A 
25 Stable disease Yes 22.78 PIK3CA E545A One copy deletion: PTEN N/A N/A N/A 
29 Stable disease  2.99 N/A High gain: AKT2, PIK3CG, PTK7 175 15 100 
Patient IDDose cohortBest responseCB18PFS (months)DNA variantsCopy-number variationPTK7 H-score (screen)PTK7 H-score (C1D15)Nuc-pAKT (screen)Cyto-pAKT (screen)Nuc-pAKT (C1D15)Cyto-pAKT (C1D15)
Progressive disease  1.12 N/A N/A 120 166 10 
Progressive disease  1.22 N/A N/A 16 25 
Progressive disease  1.15 N/A High gain: AKT2, PIK3CB N/A 232 N/A N/A 
Progressive disease  2.1 PTK7 E139Q High gain: AKT2, PIK3CB 101 22 
Progressive disease  0.69 AKT2 S268L N/A 197 N/A 20 N/A N/A 
Partial response Yes 11.31 PTEN I101N N/A 116 N/A 200 10 
Progressive disease  1.22 N/A N/A 224 116 10 
10 Progressive disease  1.84 N/A N/A 68 82 
12 Progressive disease  1.51 N/A N/A 111 N/A 60 N/A N/A 
15 Partial response Yes 6.21 N/A High gain: AKT2 144 N/A 20 N/A N/A 
16 Progressive disease  1.45 PIK3CA H1047R High gain: AKT2 168 N/A 140 N/A N/A 
18 Not evaluable  — N/A N/A N/A N/A N/A N/A N/A N/A 
19 Partial response Yes 7.36 NS NS N/A N/A N/A N/A N/A N/A 
20 Stable disease Yes 7.07 N/A High gain: AKT2 120 120 200 N/A N/A 
21 Progressive disease  3.02 NS NS N/A N/A N/A N/A N/A N/A 
22 Not evaluable  — NS NS N/A N/A N/A N/A N/A N/A 
25 Stable disease Yes 22.78 PIK3CA E545A One copy deletion: PTEN N/A N/A N/A 
29 Stable disease  2.99 N/A High gain: AKT2, PIK3CG, PTK7 175 15 100 

Note: DNA aberrations and copy-number variation of focus were in the PI3K pathway (PIK3CA, PIK3CB, PIK3CD, PIK3CG, AKT1, AKT2, AKT3, PTEN) and PTK7.

Abbreviations: N/A, samples did not have mutations in the PI3K or PTK7 pathway; NS, not sequenced.

Of note, in our pt with exceptional response to therapy (Pt #: 0613-25), we observed that the pt had a concurrent PIK3CA E545A gain-of-function mutation and a PTEN deletion. These data demonstrate two mutations that are consistent with potential activation of the PI3K pathway and may help explain the exceptional response seen in this pt.

Tissues from research biopsies obtained prior to therapy and at C1D15 were stained for the expression of PTK7 as well as pAKT, a pharmacodynamic marker of PI3K inhibition (Table 4). There was no correlation between IHC H-scores for either marker and timepoint with clinical benefit. However, due to the small sample size, this observation is preliminary.

In this study, we have shown that the novel combination of gedatolisib and cofetuzumab pelidotin can be delivered in combination at full doses with manageable toxicity and has clinical activity in pts with heavily pretreated metastatic TNBC. While the PI3K pathway is genomically and transcriptomically aberrant in the majority of TNBCs, single-agent treatment with inhibitors of this pathway has resulted in only modest clinical activity, contrary to the theory of oncogene addiction (34, 35). Transcriptome reprogramming has emerged as a common response mechanism in tumor cells when exposed to small-molecule perturbations (36). Four independent groups have reported that upregulation of the Wnt pathway is a common compensatory response to PI3K inhibition driving resistance (19–22). Inhibition of both pathways resulted in synergistic antitumor efficacy in preclinical models (19). We theorized that we could leverage this compensatory mechanism to inform clinical combination therapy by using a PI3K inhibitor (gedatolisib) to potentially drive upregulation of PTK7, such that synergy would be achieved with cotreatment with a PTK7-ADC (cofetuzumab pelidotin). Furthermore, previously reported data has also shown that the payload for cofetuzumab pelidotin, auristatin, is in itself synergistic with gedatolisib, providing another potential mechanism for synergistic action of this combination (28).

In this phase I trial, we report the first clinical experience of combining a PI3K pathway inhibitor with an agent that targets a component of the Wnt pathway. The trial successfully completed its dose escalation to RP2D of both agents with no DLTs. Safety data from the dose escalation and our expansion cohort found the combination to be generally well tolerated. Furthermore, while our sample size was limited, we observed clinical activity of the combination with a CB18 of 27.8%. Given the paucity of targeted agents for the majority of patients with TNBC, this combination provides a promising avenue for increasing the clinical armamentarium. Our study was not powered to definitely detect biomarkers of response; however, patients with genomic aberrations in the PI3K pathway or PTK7 may preferentially benefit. Taken together, these data set the stage for larger phase II studies of the combination in metastatic TNBC.

M. Radovich reports other support from LifeOmic, Tyme Technologies, and Caris Life Sciences; personal fees and non-financial support from Eli Lilly; and grants from Boston Biomedical outside the submitted work. J.P. Solzak reports other support from Caris Life Sciences, Inc. outside the submitted work. C.J. Wang reports other support from Caris Life Sciences outside the submitted work. B.A. Hancock reports other support from LifeOmic outside the submitted work. S. Badve reports grants from Agilent and Lilly, as well as personal fees from BMS outside the submitted work. S.M. Bray reports other support from LifeOmic, Inc. outside the submitted work. T.J. Ballinger reports personal fees from Medscape and Novartis outside the submitted work. B.P. Schneider reports non-financial support from Genentech, Pfizer, Foundation Medicine, and Epic Sciences, as well as personal fees from Eli Lilly outside the submitted work. K.D. Miller reports non-financial support from Pfizer, as well as grants from Breast Cancer Research Foundation during the conduct of the study. K.D. Miller also reports personal fees from Merck, Roche/Genentech, and AstraZeneca, as well as grants from Astex, Pfizer, CytoMx, AbbVie, Amgen, British Biotech, Merck, and Seattle Genetics outside the submitted work. No disclosures were reported by the other authors.

M. Radovich: Conceptualization, resources, formal analysis, supervision, funding acquisition, investigation, methodology, writing–original draft, project administration, writing–review and editing. J.P. Solzak: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. C.J. Wang: Investigation, writing–review and editing. B.A. Hancock: Investigation, writing–review and editing. S. Badve: Investigation. S.K. Althouse: Data curation, formal analysis, visualization, writing–review and editing. S.M. Bray: Data curation, software, visualization, writing–review and editing. A.M.V. Storniolo: Investigation, writing–review and editing. T.J. Ballinger: Investigation, writing–review and editing. B.P. Schneider: Investigation, writing–review and editing. K.D. Miller: Conceptualization, resources, formal analysis, supervision, funding acquisition, investigation, methodology, writing–original draft, project administration, writing–review and editing.

This work was supported by the NIH/NCI R21CA229951 (M. Radovich); Breast Cancer Research Foundation (K.D. Miller); 100 Voices of Hope; The Vera Bradley Foundation for Breast Cancer Research; The Catherine Peachey Fund; and the Indiana University Precision Health Initiative. Drug support was provided by Pfizer.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Dent
R
,
Trudeau
M
,
Pritchard
KI
,
Hanna
WM
,
Kahn
HK
,
Sawka
CA
, et al
.
Triple-negative breast cancer: clinical features and patterns of recurrence
.
Clin Cancer Res
2007
;
13
:
4429
34
.
2.
Schneider
BP
,
Winer
EP
,
Foulkes
WD
,
Garber
J
,
Perou
CM
,
Richardson
A
, et al
.
Triple-negative breast cancer: risk factors to potential targets
.
Clin Cancer Res
2008
;
14
:
8010
8
.
3.
Pal
SK
,
Childs
BH
,
Pegram
M
.
Triple negative breast cancer: unmet medical needs
.
Breast Cancer Res Treat
2011
;
125
:
627
36
.
4.
Carey
L
,
Winer
E
,
Viale
G
,
Cameron
D
,
Gianni
L
.
Triple-negative breast cancer: disease entity or title of convenience?
Nat Rev Clin Oncol
2010
;
7
:
683
92
.
5.
Liedtke
C
,
Mazouni
C
,
Hess
KR
,
Andre
F
,
Tordai
A
,
Mejia
JA
, et al
.
Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer
.
J Clin Oncol
2008
;
26
:
1275
81
.
6.
Schmid
P
,
Cortes
J
,
Pusztai
L
,
McArthur
H
,
Kummel
S
,
Bergh
J
, et al
.
Pembrolizumab for early triple-negative breast cancer
.
N Engl J Med
2020
;
382
:
810
21
.
7.
Cortes
J
,
Cescon
DW
,
Rugo
HS
,
Nowecki
Z
,
Im
SA
,
Yusof
MM
, et al
.
Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for previously untreated locally recurrent inoperable or metastatic triple-negative breast cancer (KEYNOTE-355): a randomised, placebo-controlled, double-blind, phase 3 clinical trial
.
Lancet
2020
;
396
:
1817
28
.
8.
Modi
S
,
D'Andrea
G
,
Norton
L
,
Yao
TJ
,
Caravelli
J
,
Rosen
PP
, et al
.
A phase I study of cetuximab/paclitaxel in patients with advanced-stage breast cancer
.
Clin Breast Cancer
2006
;
7
:
270
7
.
9.
Modi
S
,
Seidman
AD
,
Dickler
M
,
Moasser
M
,
D'Andrea
G
,
Moynahan
ME
, et al
.
A phase II trial of imatinib mesylate monotherapy in patients with metastatic breast cancer
.
Breast Cancer Res Treat
2005
;
90
:
157
63
.
10.
O'Shaughnessy
J
,
Weckstein
DJ
,
Vukelja
SJ
,
McIntyre
K
,
Krekow
L
,
Holmes
FA
, et al
.
Preliminary results of a randomized phase II study of weekly irinotecan/carboplatin with or without cetuximab in patients with metastatic breast cancer
.
Breast Cancer Res Treat
2007
;
106
:
S32
3
.
11.
Baselga
J
,
Albanell
J
,
Ruiz
A
,
Lluch
A
,
Gascon
P
,
Guillem
V
, et al
.
Phase II and tumor pharmacodynamic study of gefitinib in patients with advanced breast cancer
.
J Clin Oncol
2005
;
23
:
5323
33
.
12.
Tutt
A
,
Robson
M
,
Garber
JE
,
Domchek
SM
,
Audeh
MW
,
Weitzel
JN
, et al
.
Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and advanced breast cancer: a proof-of-concept trial
.
Lancet
2010
;
376
:
235
44
.
13.
Fong
PC
,
Boss
DS
,
Yap
TA
,
Tutt
A
,
Wu
P
,
Mergui-Roelvink
M
, et al
.
Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers
.
N Engl J Med
2009
;
361
:
123
34
.
14.
Kalinsky
K
,
Diamond
JR
,
Vahdat
LT
,
Tolaney
SM
,
Juric
D
,
O'Shaughnessy
J
, et al
.
Sacituzumab govitecan in previously treated hormone receptor-positive/HER2-negative metastatic breast cancer: final results from a phase I/II, single-arm, basket trial
.
Ann Oncol
2020
;
31
:
1709
18
.
15.
Bardia
A
,
Tolaney
SM
,
Loirat
D
,
Punie
K
,
Oliveira
M
,
Rugo
HS
, et al
.
ASCENT: A randomized phase III study of sacituzumab govitecan vs treatment of physician's choice in patients with previously treated metastatic triple-negative breast cancer (mTNBC)
.
Ann Oncol
2020
;
31
:
S1142
1215
.
16.
Radovich
M
,
Clare
SE
,
Atale
R
,
Pardo
I
,
Hancock
BA
,
Solzak
JP
, et al
.
Characterizing the heterogeneity of triple-negative breast cancers using microdissected normal ductal epithelium and RNA-sequencing
.
Breast Cancer Res Treat
2014
;
143
:
57
68
.
17.
Comprehensive molecular portraits of human breast tumours
.
Nature
2012
;
490
:
61
70
.
18.
Garrido-Castro
AC
,
Saura
C
,
Barroso-Sousa
R
,
Guo
H
,
Ciruelos
E
,
Bermejo
B
, et al
.
Phase 2 study of buparlisib (BKM120), a pan-class I PI3K inhibitor, in patients with metastatic triple-negative breast cancer
.
Breast Cancer Res
2020
;
22
:
120
.
19.
Solzak
JP
,
Atale
RV
,
Hancock
BA
,
Sinn
AL
,
Pollok
KE
,
Jones
DR
, et al
.
Dual PI3K and Wnt pathway inhibition is a synergistic combination against triple negative breast cancer
.
NPJ Breast Cancer
2017
;
3
:
17
.
20.
Tzeng
HE
,
Yang
L
,
Chen
K
,
Wang
Y
,
Liu
YR
,
Pan
SL
, et al
.
The pan-PI3K inhibitor GDC-0941 activates canonical WNT signaling to confer resistance in TNBC cells: resistance reversal with WNT inhibitor
.
Oncotarget
2015
;
6
:
11061
73
.
21.
Zhong
Z
,
Sepramaniam
S
,
Chew
XH
,
Wood
K
,
Lee
MA
,
Madan
B
, et al
.
PORCN inhibition synergizes with PI3K/mTOR inhibition in Wnt-addicted cancers
.
Oncogene
2019
;
38
:
6662
77
.
22.
Park
YL
,
Kim
HP
,
Cho
YW
,
Min
DW
,
Cheon
SK
,
Lim
YJ
, et al
.
Activation of WNT/beta-catenin signaling results in resistance to a dual PI3K/mTOR inhibitor in colorectal cancer cells harboring PIK3CA mutations
.
Int J Cancer
2019
;
144
:
389
401
.
23.
Del Campo
JM
,
Birrer
M
,
Davis
C
,
Fujiwara
K
,
Gollerkeri
A
,
Gore
M
, et al
.
A randomized phase II non-comparative study of PF-04691502 and gedatolisib (PF-05212384) in patients with recurrent endometrial cancer
.
Gynecol Oncol
2016
;
142
:
62
9
.
24.
Shapiro
GI
,
Bell-McGuinn
KM
,
Molina
JR
,
Bendell
J
,
Spicer
J
,
Kwak
EL
, et al
.
First-in-human study of PF-05212384 (PKI-587), a small-molecule, intravenous, dual inhibitor of PI3K and mTOR in patients with advanced cancer
.
Clin Cancer Res
2015
;
21
:
1888
95
.
25.
Damelin
M
,
Bankovich
A
,
Bernstein
J
,
Lucas
J
,
Chen
L
,
Williams
S
, et al
.
A PTK7-targeted antibody-drug conjugate reduces tumor-initiating cells and induces sustained tumor regressions
.
Sci Transl Med
2017
;
9
:
eaag2611
.
26.
Tolcher
AW
,
Calvo
E
,
Doger
B
,
Maitland
ML
,
Gibson
B
,
Xuan
D
, et al
.
A phase I study of PF-06647020, an antibody-drug conjugate targeting protein kinase 7 (PTK7), in patients with advanced solid tumors
;
2015
.
27.
Lu
X
,
Borchers
AG
,
Jolicoeur
C
,
Rayburn
H
,
Baker
JC
,
Tessier-Lavigne
M
.
PTK7/CCK-4 is a novel regulator of planar cell polarity in vertebrates
.
Nature
2004
;
430
:
93
8
.
28.
Shor
B
,
Kahler
J
,
Dougher
M
,
Xu
J
,
Mack
M
,
Rosfjord
E
, et al
.
Enhanced antitumor activity of an anti-5T4 antibody-drug conjugate in combination with PI3K/mTOR inhibitors or taxanes
.
Clin Cancer Res
2016
;
22
:
383
94
.
29.
Jackson-Fisher
A
,
Mehra
N
,
Gianani
R
,
Whalen
P
,
Vizcarra
P
,
Deng
S
, et al
.
Protein tyrosine kinase 7 (PTK7) biomarker analysis in patients (pts) treated with PF-06647020, a PTK7 antibody-drug conjugate (ADC), in a phase I dose expansion study [abstract]
. In:
Proceedings of the American Association for Cancer Research Annual Meeting 2019
;
2019 Mar 29–Apr 3
;
Atlanta, GA. Philadelphia (PA)
:
AACR
;
Cancer Res
2019
;
79
(
13 Suppl
):
Abstract nr 4035
.
30.
Badve
S
,
Collins
NR
,
Bhat-Nakshatri
P
,
Turbin
D
,
Leung
S
,
Thorat
M
, et al
.
Subcellular localization of activated AKT in estrogen receptor- and progesterone receptor-expressing breast cancers: potential clinical implications
.
Am J Pathol
2010
;
176
:
2139
49
.
31.
Li
H
,
Durbin
R
.
Fast and accurate short read alignment with Burrows-Wheeler transform
.
Bioinformatics
2009
;
25
:
1754
60
.
32.
Van der Auwera
GA
,
Carneiro
MO
,
Hartl
C
,
Poplin
R
,
Del Angel
G
,
Levy-Moonshine
A
, et al
.
From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline
.
Curr Protoc Bioinformatics
2013
;
43
:
11.10.1–11.10.33
.
33.
Jiang
Y
,
Wang
R
,
Urrutia
E
,
Anastopoulos
IN
,
Nathanson
KL
,
Zhang
NR
.
CODEX2: full-spectrum copy number variation detection by high-throughput DNA sequencing
.
Genome Biol
2018
;
19
:
202
.
34.
Weinstein
IB
,
Joe
AK
.
Mechanisms of disease: Oncogene addiction–a rationale for molecular targeting in cancer therapy
.
Nat Clin Pract Oncol
2006
;
3
:
448
57
.
35.
Weinstein
IB
,
Joe
A
.
Oncogene addiction
.
Cancer Res
2008
;
68
:
3077
80
.
36.
Subramanian
A
,
Narayan
R
,
Corsello
SM
,
Peck
DD
,
Natoli
TE
,
Lu
X
, et al
.
A next generation connectivity map: L1000 platform and the first 1,000,000 profiles
.
Cell
2017
;
171
:
1437
52
.

Supplementary data