Paclitaxel plus Eftilagimod Alpha, a Soluble LAG-3 Protein, in Metastatic, HR+ Breast Cancer: Results from AIPAC, a Randomized, Placebo Controlled Phase IIb Trial

PURPOSE

Eftilagimod alpha (efti), a soluble lymphocyte activation gene (LAG-3) protein and MHC class II agonist, enhances innate and adaptive immunity. Active Immunotherapy PAClitaxel (AIPAC) evaluated safety and efficacy of efti plus paclitaxel in patients with predominantly endocrine-resistant, hormone receptor-positive, HER2-negative metastatic breast cancer (ET-resistant HR+ HER2- MBC).

PATIENTS AND METHODS

Women with HR+ HER2- MBC were randomized 1:1 to weekly intravenous paclitaxel (80 mg/m2) and subcutaneous efti (30 mg) or placebo every 2 weeks for six 4-week cycles, then monthly subcutaneous efti (30 mg) or placebo maintenance. Primary endpoint was progression-free survival (PFS) by blinded independent central review. Secondary endpoints included overall survival (OS), safety/tolerability, pharmacokinetics/pharmacodynamics, and quality of life. Exploratory endpoints included cellular biomarkers.

RESULTS

114 patients received efti and 112 patients received placebo. Median age was 60 years (91.6% visceral disease, 84.1% ET-resistant, 44.2% with previous CDK4/6 inhibitor treatment). Median PFS at 7.3 months was similar for efti and placebo. Median OS was not significantly improved for efti (20.4 vs. 17.5 months; HR, 0.88; P = 0.197) but became significant for predefined exploratory subgroups. EORTC QLQC30-B23 global health status was sustained for efti but deteriorated for placebo. Efti increased absolute lymphocyte, monocyte and secondary target cell (CD4, CD8) counts, plasma IFNγ and CXCL10 levels.

CONCLUSIONS

Although the primary endpoint, PFS, was not met, AIPAC confirmed expected pharmacodynamic effects and demonstrated excellent safety profile for efti. OS was not significantly improved globally (2.9-month difference), but was significantly improved in exploratory biomarker subgroups, warranting further studies to clarify efti's role in patients with ET-resistant HER2- MBC.

Successful Trastuzumab-Deruxtecan Rechallenge After Interstitial Lung Disease: A Case Report

Trastuzumab deruxtecan (T-DXd) is used to treat human epidermal growth factor receptor 2-positive advanced breast cancer. Interstitial lung disease (ILD) is a severe adverse event associated with T-DXd. Current guidelines recommend permanent discontinuation of T-DXd after Common Terminology Criteria for Adverse Events (CTCAE) grade ≥ 2 ILD. Here, we describe a case of successful rechallenge with T-DXd after CTCAE grade 2 treatment-induced ILD. After discontinuation of T-DXd, ILD was treated with steroids until complete resolution. Given the initial beneficial antitumor response, retreatment was discussed during disease progression. In a shared decision with the patient, T-DXd was restarted at the lowest registered dose, along with low-dose steroids. ILD did not reoccur. Importantly, both clinical and radiological responses to the treatment were observed, with an improvement in the patient's quality of life. This case demonstrates that retreatment with T-DXd after a grade 2 ILD event is feasible and yields clinical benefit.

Explainable artificial intelligence (XAI) in radiology and nuclear medicine: a literature review

Rational

Deep learning (DL) has demonstrated a remarkable performance in diagnostic imaging for various diseases and modalities and therefore has a high potential to be used as a clinical tool. However, current practice shows low deployment of these algorithms in clinical practice, because DL algorithms lack transparency and trust due to their underlying black-box mechanism. For successful employment, explainable artificial intelligence (XAI) could be introduced to close the gap between the medical professionals and the DL algorithms. In this literature review, XAI methods available for magnetic resonance (MR), computed tomography (CT), and positron emission tomography (PET) imaging are discussed and future suggestions are made.

Methods

PubMed, Embase.com and Clarivate Analytics/Web of Science Core Collection were screened. Articles were considered eligible for inclusion if XAI was used (and well described) to describe the behavior of a DL model used in MR, CT and PET imaging.

Results

A total of 75 articles were included of which 54 and 17 articles described post and ad hoc XAI methods, respectively, and 4 articles described both XAI methods. Major variations in performance is seen between the methods. Overall, post hoc XAI lacks the ability to provide class-discriminative and target-specific explanation. Ad hoc XAI seems to tackle this because of its intrinsic ability to explain. However, quality control of the XAI methods is rarely applied and therefore systematic comparison between the methods is difficult.

Conclusion

There is currently no clear consensus on how XAI should be deployed in order to close the gap between medical professionals and DL algorithms for clinical implementation. We advocate for systematic technical and clinical quality assessment of XAI methods. Also, to ensure end-to-end unbiased and safe integration of XAI in clinical workflow, (anatomical) data minimization and quality control methods should be included.

Abstract PS12-04: Rintodestrant (G1T48), an oral selective estrogen receptor degrader in ER+/HER2- locally advanced or metastatic breast cancer: Updated phase 1 results and dose selection

Background: Rintodestrant (G1T48) is a potent oral selective estrogen receptor degrader (SERD) that competitively binds to the estrogen receptor (ER) and blocks ER signaling in tumors resistant to other endocrine therapies. Preliminary results from Part 1 dose escalation showed robust target engagement on 18F-fluoroestradiol positron emission tomography (FES-PET), a favorable safety profile, and encouraging antitumor activity in patients with heavily pretreated ER+/HER2- advanced breast cancer (ABC), including those with ESR1 mutations (Dees et al., ESMO 2019 [abstract #3587]). Here, we present updated results from dose escalation and expansion (Parts 1 and 2). Methods: This Phase 1, first-in-human, open-label study evaluated rintodestrant monotherapy in women with ER+/HER2- ABC after progression on endocrine therapy. Part 1 was a 3+3 dose escalation (200-1000 mg once daily [QD]); Part 2 expanded 600 and 1000 mg QD; and Part 3 was added to assess rintodestrant with palbociclib in patients in earlier lines in the advanced setting. Primary objectives included dose-limiting toxicities (DLTs), maximum tolerated dose (MTD), safety, and recommended Phase 2 dose. Secondary objectives included pharmacokinetics and antitumor activity (RECIST v1.1). Exploratory objectives included pharmacodynamic inhibition of ER target engagement (FES-PET), mutation profiling (cell-free DNA [cfDNA]), and change in ER expression from baseline to on-treatment tumor biopsies. Results: As of May 13, 2020, 67 patients (Part 1: n = 26; Part 2: n = 41) were treated, with a median age of 61 years (range 34-83) and ECOG PS of 0 (49%) or 1 (51%). Median number of prior lines in the advanced setting was 2 (range 0-9), including prior fulvestrant (64%), CDK4/6 inhibitor (69%), mTOR inhibitor (22%), and/or chemotherapy (46%). Median number of prior lines of endocrine therapy in the advanced setting was 2 (range 0-5), with 61% of patients having received ≥2 lines. Treatment-related adverse events (TRAEs) were reported in 70% of patients. The most common TRAEs in ≥10% of patients included hot flush (24%), fatigue (21%), nausea (19%), diarrhea (18%), and vomiting (10%), mostly grade 1 or 2. No DLTs were reported and MTD was not reached. Dose reduction due to TRAEs occurred in 1 patient (1%), with elevated transaminases (grade 3 ALT and grade 2 AST) at 600 mg. Serious TRAEs occurred in 2 patients at 1000 mg (grade 5 cerebral hemorrhage in the setting of low molecular weight heparin and grade 2 upper abdominal pain). Two patients (3%) discontinued treatment due to TRAEs. Overall, the frequency of patients with TRAEs at 800 mg was comparable with that at 600 mg (57% vs 63%) and less than that at 1000 mg (81%). Of 67 patients, 16 were on study treatment for ≥24 weeks and 3 (n = 1 at 600 mg; n = 2 at 1000 mg, including 1 with ESR1 mutation) had a confirmed partial response (clinical benefit rate [CBR]: 28%). FES-PET standard uptake values decreased at week 4 with a mean reduction of 87% (±8%) at doses ≥ 600 mg. Of 59 patients tested for baseline cfDNA, 41% harbored ≥1 ESR1 mutation, with a similar CBR in both groups (33% in ESR1 mutant and 29% in ESR1 wild-type). Seven of 9 patients had a decrease in ER immunohistochemistry H-score at both 600 and 1000 mg (median [range]: -27.8% [-33.8%, -3.4%]), irrespective of ESR1 mutation status. Based on safety, efficacy, and ER degradation, 800 mg was selected as the optimal dose for further study. Conclusions: Rintodestrant continues to demonstrate an excellent safety/tolerability profile across all doses, with promising antitumor activity in patients with heavily pretreated ER+/HER2- ABC, including those with tumors harboring ESR1 mutations. Part 3 of this study, evaluating rintodestrant 800 mg QD with palbociclib in a more endocrine-sensitive population, is ongoing (NCT03455270).

Citation Format: Philippe Aftimos, Patrick Neven, Mark Pegram, Catharina Willemien Menke-van der Houven van Oordt, E. Claire Dees, Carolien Schröder, Agnes Jager, Iurie Bulat, Linnea Chap, Marina Maglakelidze, Erika Hamilton, Massimo Cristofanilli, Susanna Ulahannan, Jorianne Boers, Ramsha Iqbal, Adrian Crijanovschi, Curt D Wolfgang, Wenli Tao, Christina Sipes, Rajesh Malik, Sarika Jain. Rintodestrant (G1T48), an oral selective estrogen receptor degrader in ER+/HER2- locally advanced or metastatic breast cancer: Updated phase 1 results and dose selection [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS12-04.

Abstract PD8-07: Pharmacodynamic analysis from a phase 1 study of rintodestrant (G1T48), an oral selective estrogen receptor degrader, in ER+/HER2- locally advanced or metastatic breast cancer

Background: Rintodestrant is an orally bioavailable, potent and selective estrogen receptor degrader (SERD) that inhibits estrogen receptor (ER) gene transcription, degrades the ER, and delays tumor proliferation in preclinical models. Preliminary results from Part 1 dose escalation (200-1000 mg once daily) demonstrated that rintodestrant has a favorable safety profile and encouraging antitumor activity in patients (pts) with heavily pretreated ER+/HER2- advanced breast cancer (ABC) (Dees et al., ESMO 2019 [abstract #3587]). Here, we report the pharmacodynamic (PD) analysis in peripheral blood and tumor biopsies from pts who received rintodestrant in Part 1 and 2 (600 and 1000 mg dose expansion) to characterize the pt population and mechanisms of response. Methods: This Phase 1, first-in-human, open-label study evaluated rintodestrant in women with ER+/HER- ABC after progression on endocrine therapy. PD analysis included inhibition of ER target engagement with 18F-fluoroestradiol positron emission tomography (FES-PET), mutational profiling (cell-free DNA [cfDNA]), and circulating tumor cell (CTC) enumeration. Tumor biopsies sampled at baseline and 6 weeks on treatment were evaluated for ER degradation (immunohistochemistry [IHC]) and proliferation (Ki67, IHC) to understand the on-target effects of rintodestrant. Results: As of May 13, 2020, 67 pts had been treated. FES-PET data were obtained in 14 pts and showed a decrease in all pts, with maximum standard uptake values (SUVmax) ranging from 70% to 98% after 4 weeks of rintodestrant monotherapy across all doses. Fifty-nine pts were tested for cfDNA at baseline; 95% (n = 56) harbored ≥1 somatic variant (median = 3 mutations per pt). Among pts with somatic variants, 41% had ESR1 mutations, with D538G being the most common (58%). Additionally, 46% and 42% of pts harbored mutations in TP53 and PIK3CA, respectively, and 10% had mutations in both ESR1 and PIK3CA. Similar clinical benefit rates were observed in wild-type vs ESR1 and/or PIK3CA mutant tumors. An analysis of change of variant allele fraction (VAF) in 55 pts between baseline and 2 weeks of treatment revealed that 58% had a decrease in mean VAF, with a decrease in ESR1 VAF in 16/20 pts that had ESR1 mutations at baseline. Furthermore, of 24 pts who had samples collected at baseline and progression, 16 (67%) developed additional variants (median [range]: 2 [1, 15]), including EGFR, ERBB2, TP53, and ESR1. CTC analysis (n = 45) showed the mean value of Epi+CD45- CTCs decreased from 2.8 cells/mL to 1.8 cells/mL after 8 weeks of treatment. Tumor biopsies were collected in 9 pts (5 received 600 mg and 4 received 1000 mg) at baseline and 6 weeks on treatment. Of the 7/9 pts that had a decrease in the ER H-score (median [range]: -27.8% [-33.8%, -3.4%]), 4 had ≥1 variant in ESR1 at baseline. Overall, 4 pts had a decrease in Ki67, with reductions mostly observed in pts who received 600 mg rintodestrant. Additional analyses, including correlations with clinical response, are ongoing and will be presented. Conclusions: Rintodestrant demonstrated robust ER target engagement on FES-PET, as well as substantial decreases in ER H-score, cfDNA VAF, and Epi+CD45- CTCs. These data, along with promising clinical benefit in pts with heavily pretreated ER+/HER2- ABC, regardless of ESR1 or PIK3CA mutation status, warrant additional investigation of rintodestrant (NCT03455270).

Citation Format: Philippe Aftimos, Marina Maglakelidze, Andor WJM Glaudemans, Erika Hamilton, Linnea Chap, Elisabeth de Vries, Catharina Willemien Menke-van der Houven van Oordt, Agnes Jager, E. Claire Dees, Massimo Cristofanilli, Mark Pegram, Susanna Ulahannan, Patrick Neven, Iurie Bulat, Ruhi Rai, Wenli Tao, Sarika Jain, Andrew P Beelen, Jessica A Sorrentino. Pharmacodynamic analysis from a phase 1 study of rintodestrant (G1T48), an oral selective estrogen receptor degrader, in ER+/HER2- locally advanced or metastatic breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD8-07.

Neutrophils Kill Antibody-Opsonized Cancer Cells by Trogoptosis

Destruction of cancer cells by therapeutic antibodies occurs, at least in part, through antibody-dependent cellular cytotoxicity (ADCC), and this can be mediated by various Fc-receptor-expressing immune cells, including neutrophils. However, the mechanism(s) by which neutrophils kill antibody-opsonized cancer cells has not been established. Here, we demonstrate that neutrophils can exert a mode of destruction of cancer cells, which involves antibody-mediated trogocytosis by neutrophils. Intimately associated with this is an active mechanical disruption of the cancer cell plasma membrane, leading to a lytic (i.e., necrotic) type of cancer cell death. Furthermore, this mode of destruction of antibody-opsonized cancer cells by neutrophils is potentiated by CD47-SIRPα checkpoint blockade. Collectively, these findings show that neutrophil ADCC toward cancer cells occurs by a mechanism of cytotoxicity called trogoptosis, which can be further improved by targeting CD47-SIRPα interactions.

89Zr-cetuximab PET imaging in patients with advanced colorectal cancer

Monoclonal antibodies (mAbs) against the epidermal growth factor receptor (EGFR) are used in the treatment of advanced colorectal cancer (mCRC). Approximately 50% of patients benefit despite patient selection for RAS wild type (wt) tumors. Based on the hypothesis that tumor targeting is required for clinical benefit of anti-EGFR treatment, biodistribution and tumor uptake of (89)Zr-cetuximab by Positron Emission Tomography (PET), combining the sensitivity of PET with the specificity of cetuximab for EGFR was evaluated. Ten patients with wt K-RAS mCRC received 37 ± 1 MBq (89)Zr-cetuximab directly (<2 h) after the first therapeutic dose of cetuximab. PET-scans were performed from 1 hour to 10 days post injection (p.i.). Biodistribution was determined for blood and organs. Uptake in tumor lesions was quantified by Standardized Uptake Value (SUV) and related to response. In 6 of 10 patients (89)Zr-cetuximab uptake in tumor lesions was detected. Four of 6 patients with (89)Zr-cetuximab uptake had clinical benefit, while progressive disease was observed in 3 of 4 patients without (89)Zr-cetuximab uptake. Taken together, tumor uptake of 89Zr-cetuximab can be visualized by PET imaging. The strong relation between uptake and response warrants further clinical validation as an innovative selection method for cetuximab treatment in patients with wt RAS mCRC.