Abstract Background: Multiple trials have proven that adding CDK4/6 inhibitors to endocrine treatment increases progression free (PFS) and overall survival (OS) of patients with metastatic estrogen receptor (ER)-positive HER2 negative breast cancer. However, only limited evidence is available of treatment efficacy of CDK4/6 inhibitors for invasive lobular carcinoma (ILC). This retrospective study aims to compare treatment duration of the three FDA/EMA approved CDK4/6 inhibitors in patients with no special type (NST) breast cancer versus patients with ILC. Methods: All patients with metastatic ER-positive HER2-negative breast cancer who were treated with a CDK4/6 inhibitor (1st, 2nd; 3rd line) in University Hospitals Leuven between December 2014 and February 2023, were included. A comparison of PFS and OS was made between patients with NST and ILC by use of the Kaplan Meier method. Other histological subtypes as well as mixed subtypes were excluded. Uni- and multivariable cox regression models were performed to quantify the association of histological subtype with PFS and OS. Results: A total of 418 patients were included of which 119 (28.5%) patients with ILC (median age at primary diagnosis 59 years, range 36 – 89 years) and 299 (71.5%) patients with NST (median age at primary diagnosis 55 years, range 23 – 90 years). Median follow-up was 26.8 months (range 1.1 – 88.1 months). Median PFS was 15.2 months (range 1.0 – 74.6 months) and 14.7 months (range 1.0 – 89.3 months) for patients with ILC and NST respectively. The OS rate after 60 months follow up was 39.2 % (CI 26.8 – 51.4) for ILC and 40.1% (CI 32.2 – 47.8) for NST. As shown in table 1, clear differences were observed in PFS rates after 12, 24 and 60 months between patients with NST that received CDK4/6 inhibitors in first line versus second or third line. These differences were less apparent for patients with ILC. For both NST and ILC, endocrine resistance at start of CDK4/6 inhibition impacted PFS rates negatively (table 2). In multivariable analyses, histological subtype was not predictive for CDK4/6 inhibition outcome (PFS: hazard ratio (HR) 0.996, CI 0.728 – 1.364, p-value 0.981; OS: HR 0.872, CI 0.602 – 1.263, p-value 0.468). Endocrine sensitivity vs. resistance at the time CDK4/6 inhibitor was started, was proven to be predictive for both PFS (HR 0.478, CI 0.337 – 0.678, p-value < 0.001) and OS (HR 0.528, CI 0.347 – 0.804, p-value 0.003). Conclusion: In our center, the histological subtype of breast cancer did not seem to impact PFS and OS significantly after treatment with CDK4/6 inhibition. Patients with NST seemed to have an increased benefit in PFS from treatment with CDK4/6 inhibition in first line as compared to later lines. While for patients with ILC, no difference between treatment lines were observed. Therefore, clinicians might be able to safely postpone CDK4/6 inhibitors to second- or third-line treatment in patients with metastatic ILC. Table 1: PFS rates by histological subtype and line of CDK4/6 treatment Table 2: PFS rates by histological subtype and endocrine sensitivity at start of CDK4/6 inhibition Citation Format: Cedric Bauters, Karen Van Baelen, Hans Wildiers, Giuseppe Floris, Sileny Han, Patrick Berteloot, Thaïs Baert, Ann Smeets, Ines Nevelsteen, Yannick Van Herck, Anne Deblander, Chantal Remmerie, Lieke Dullens, Maxime Van Houdt, Annouschka Laenen, Christine Desmedt, Patrick Neven. The effect of histological breast cancer subtype on progression free survival in patients with CDK4/6 inhibitors [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-05-05.

Abstract Introduction: Currently, patients with luminal metastatic breast cancer receive a CDK4/6 inhibitor in combination with endocrine therapy as a first or second line of treatment. When tumors become resistant to the CDK4/6 inhibitors, treatment with alpelisib (PI3K inhibitor) in combination with an endocrine agent can be a next treatment option if an activating PIK3CA mutation is confirmed (cfr. SOLAR-1 trial). However, limited data is available after treatment with CDK4/6 inhibitors (cfr. BYLieve trial) and even less about alpelisib in later lines of therapy. Therefore, we aim to investigate the therapeutic efficacy and safety of alpelisib in combination with an endocrine agent in PIK3CA-mutated advanced breast cancer patients in later lines after prior CDK4/6i treatment. Methods: This is an open-label, prospective, multi-centre, single arm clinical trial enrolling patients from both the University Hospital of Leuven (UHL) and Ghent University Hospital (GUH). For each patient, metastatic tumor was tested for PIK3CA mutations using the UHL 96 gene panel. Our endpoints are clinical benefit rate, progression-free survival, time to treatment discontinuation (defined as the date of starting alpelisib to the date of treatment discontinuation or death) and safety. Descriptive statistics were used. Results: Between June 18th 2019 and August 23rd 2021, 38 patients have been included with confirmed PIK3CA hotspot mutations. All patients had CDK4/6 inhibitor in an earlier treatment line. Median age at alpelisib initiation was 64 years (range: 39-82 years). Included patients had a median of four lines of systemic therapy for advanced disease prior to starting alpelisib (range: 1- 11). Clinical benefit rate (patients receiving ³ 6 months of treatment) was 26.3% (10/38). The median progression-free-survival as well as time to treatment discontinuation were 3 months (range: 1 – 18). Dose reduction due to toxicity occurred in 11 patients (29%), of which 7 due to hyperglycemia grade 3, 2 due to diarrhea grade 3, 1 due to rash grade 2 and 1 due to anorexia. Finally, 84% of patients (32/38) stopped alpelisib treatment because of progressive disease, 8% (3/38) because of intolerance, one patient died because of cerebral hemorrhage during treatment, one patient withdrew their informed consent and one patient is still ongoing with the treatment. Conclusion: Our trial demonstrates activity of alpelisib in patients with PIK3CA mutated advanced breast cancer in later lines of treatment after treatment with CDK4/6 inhibitors with a clinical benefit rate of 26.3%. Discontinuation of therapy due to toxicity was seen in only 8% of patients, although toxicity induced dose reduction was needed in 29% of patients. These findings support the results of the BYLieve trial and additionally show efficacy of alpelisib in later lines of treatment for hormone receptor-positive advanced breast cancer. Citation Format: Rik Van Severen, Anne-Sofie De Crem, Hava Izci, Laurence Slembrouck, Isabelle Vanden Bempt, Hans Wildiers, Kevin Punie, Eline Naert, Ingeborg Hilderson, Ann Smeets, Ines Nevelsteen, Anne Deblander, Nynke Willers, Patrick Berteloot, Ignace Vergote, Sileny Han, Adriaan Vanderstichele, Giuseppe Floris, Christine Desmedt, Hannelore Denys, Patrick Neven. Efficacy and Safety of Alpelisib in PIK3CA-mutated, hormone receptor-positive advanced breast cancer after a CDK4/6 inhibitor: An Open-label, Multi-centre, Prospective, Single Arm Clinical Trial [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-21-10.

Introduction/Background PAOLA-1/ENGOTov25 trial showed PFS and OS benefit with PARP inhibitor (PARPi) and bevacizumab in maintenance treatment of BRCAm and homologous recombination deficient (HRD+) ovarian cancer. However, current available HRD tests, including our earlier reported Leuven HRD test (EJC,2023;188:131–9), can still be improved. We developed the new 'Leuven PARPi Benefit Test' as alternative approach to predict benefit of PARPi, and compared our test with the Myriad myChoice CDxPLUS test (Myriad test) in a training and validation cohort of the PAOLA-1 trial. Methodology Leuven PARPi Benefit Test is based on genome-wide SNPs of the Leuven HRD test and mutation detection of BRCA1/2 coding exons sequenced with a custom-made capture panel. A random forest-based model to identify patients that benefit most of PARPi was used to define an algorithm combining loss of heterozygosity detection with additional copy-number related features and information on functional domains of BRCA variants. The model was trained on 198 samples and validated on 198 separate samples from PAOLA-1. Results In the validation cohort, the Leuven PARPi Benefit Test classified 47.5%(94/198) samples as positive compared to 56.6%(112/198) with the Myriad test. At 2 years, PFS of Leuven PARPi Benefit Test positive patients was 78.1% vs 32.1% with olaparib vs placebo, respectively (HR 0.32;95% CI 0.181—0.566). For Myriad HRD test positive patients, this was 69.2% vs 34.4% with olaparib vs placebo, respectively(HR0.40;0.248—0.661). For patients with Leuven PARPi Benefit Test positive/BRCAwt tumours, PFS was 70.8% vs 22.2% at 2y (HR 0.18; 0.068—0.487). This was 53.6% vs 29.4% (HR 0.35;0.173—0.700) for patients with Myriad test positive/BRCAwt tumours. There was no difference in 2yPFS with either a negative Leuven PARPi Benefit Test or a negative Myriad HRD test. Conclusion Leuven PARPi Benefit Test is an improved approach to detect benefit of PARPi in ovarian cancer patients suggesting a better predictive value compared with the Myriad test. Disclosures No financial disclosures, patents pending.

Bland-Altman plot of endoxifen (in µg/l) at 3 months and at 6 months.

Background The Regan Composite Risk Score (RCRS) is a web-based prognostic and predictive calculator to guide the use of adjuvant exemestane plus ovarian function suppression (AI + OFS) versus tamoxifen plus ovarian function suppression (TAM + OFS) or tamoxifen alone (TAM) for premenopausal women with hormone receptor-positive HER2-negative early breast cancer (HR+/HER2- EBC). We compared our adjuvant endocrine therapy policy based on the tumor board with the treatment guided by the RCRS during 2 time periods, one before and one after the acquaintance of the Tamoxifen and Exemestane Trial (TEXT) and Suppression and Ovarian Function Trial (SOFT) data. This allowed us to see a possible evolution in therapy policy. Methods A retrospective cohort study of 563 premenopausal patients with HR+/HER2- and HER2+ EBC diagnosed at the University Hospital of Leuven during 2 periods, 2010-2012 (cohort 1) and 2015-2017 (cohort 2), was conducted. For each patient with HER2- EBC, the RCRS was calculated by entering the requested characteristics in the online available tool. The primary outcome was to investigate how frequent our therapy differed from the therapy guided by the RCRS based on the estimated 8-yr distant relapse free interval (DRFI) with an arbitrary cut-off set at 3 %. If the received therapy was ≥ 3 % less efficient in 8-year DRFI compared to the optimal therapy according to RCRS, the patient was considered undertreated. If the received therapy differed by less than 3 % in 8-year DRFI compared to the optimal therapy according to RCRS and yet the most intensive therapy (AI + OFS > TAM + OFS > TAM) was administered, the patient was considered overtreated. In the other cases, the patient was considered to have been treated concordant with the RCRS. Secondarily, nonadherence of the HER2- and HER2+ patients towards the endocrine treatments leading to therapy switch because of intolerance was recorded at 6, 12, 24 and 36 months. Analyses were performed using SAS software and the comparison of both cohorts was performed by the chi-squared test for categorical variables. Results According to the RCRS, 43.2 % (89/206) of the HER2-negative patients of cohort 1 were undertreated compared to 22.1 % (43/194) in cohort 2 (chi- squared test, p-value < 0.001). The number of overtreated patients also differed significantly between the two cohorts (chi-squared test, p-value = 0.003) with 2.9 % (6/206) in the first cohort and 10.3 % (20/194) in the second cohort. Finally, the number of patients treated concordant with the guidance derived from the RCRS was 53.9 % (111/206) in cohort 1 and 67.5 % (131/194) in cohort 2 (chi-squared test, p-value = 0.005). Treatment intolerance and switch was observed in 34.8 %, 16.7 % and 12.4 % of the patients receiving AI + OFS, TAM + OFS or TAM as initial therapy respectively; this was numerically higher for all treatments in cohort 2 vs cohort 1, although the observed difference was only significant for TAM. Conclusion In our center, a recent cohort of premenopausal women was more likely to be treated with the adjuvant endocrine treatment concordant with the guidance derived from the RCRS when using an arbitrary cut-off of 3 % to define a relevant improvement in outcome. Citation Format: Charlotte Berteloot, Patrick Neven, Maja Vangoitsenhoven, Annouschka Laenen, Hans Wildiers, Kevin Punie, Ann Smeets, Ines Nevelsteen, Sileny Han, Thaïs Baert, Hilde Janssen, Eva Oldenburger, Adinda Baten, Patrick Berteloot, Rani Vanhoudt, Anne Deblander, Chantal Remmeriev, Christine Desmedt. Real world adjuvant endocrine treatment in premenopausal breast cancer patients compared with the proposed algorithm using the Regan Composite Risk Score [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P1-02-04.

Kaplan-Meier curves stratified by endoxifen levels using a quartile split.

Results of ordinal regression models of the questionnaire items on endoxifen level after multiple imputation.

BACKGROUND:The PAOLA-1/ENGOT-ov25 trial showed improved progression-free (PFS) and overall survival (OS) in homologous recombination deficient (HRD) positive patients treated with olaparib, but not when HRD negative (HRD tested with MyChoice CDx PLUS [Myriad test]). PATIENTS AND METHODS:The academic Leuven HRD test consists of capture-based targeted sequencing of genome-wide single-nucleotide polymorphisms and coding exons of eight HR genes including BRCA1, BRCA2, and TP53. We compared the predictive value of the Leuven HRD versus Myriad HRD test for PFS and OS in the randomised PAOLA-1 trial. RESULTS:468 patients had left-over DNA after Myriad testing for Leuven HRD testing. Positive/negative/overall percent agreement for the Leuven versus Myriad HRD status was 95%/86%/91%, respectively. Tumours were HRD+ in 55% and 52%, respectively. In Leuven HRD+ patients, 5years PFS (5yPFS) was 48.6% versus 20.3% (HR 0.431; 95% confidence intervals (CI) 0.312-0.595) for olaparib versus placebo, respectively (Myriad test 0.409; 95% CI 0.292-0.572). In Leuven HRD+/BRCAwt patients 5yPFS was 41.3% versus 12.6% (HR 0.497; 95% CI 0.316-0.783), and 43.6% versus 13.3% (HR 0.435; 95% CI 0.261-0.727) for the Myriad test. 5yOS was prolonged in the HRD+ subgroup with both tests 67.2% versus 54.4% (HR 0.663; 95% CI 0.442-0.995) for the Leuven test, and 68.0% versus 51.8% (HR 0.596 95% CI 0.393-0.904) for the Myriad test. HRD status was undetermined in 10.7% and 9.4% of the samples, respectively. CONCLUSIONS:A robust correlation between the Leuven HRD and Myriad test was observed. For HRD+ tumours, the academic Leuven HRD showed a similar difference in PFS and OS as the Myriad test.

Flowchart of patients who were included/excluded from the study.

Auto-immuunaandoeningen hebben een zeer uiteenlopende presentatie en de interpretatie van auto-immuunserologie is, gezien de vele valkuilen, een uitdaging. Deze factoren maken het stellen van een correcte diagnose erg complex. Dit e-book bundelt daarom de belangrijkste artikels, die in 2021-2022 werden gepubliceerd in het Tijdschrift voor Geneeskunde en Gezondheidszorg rond dit thema. De auteurs reiken u […]

Objectives: The phase III PAOLA-1/ENGOT-ov25 trial (NCT02477644) evaluated first-line standard therapy including bevacizumab in advanced ovarian cancer with the addition of maintenance olaparib or placebo. Significant improved progression-free survival (PFS) was observed in homologous recombination deficient (HRD) tumors, with or without BRCA mutation (BRCAm), tested with Myriad myChoice PLUS (Myriad test) in contradiction to homologous recombination proficient (HRP) tumors, revealing the need for HRD testing in first line. As part of the ENGOT European HRD initiative, we developed the ‘Leuven’ HRD test. To the best of our knowledge, this is the first presentation of an alternative academic laboratory-developed HRD test compared with the Myriad test including an analysis of the predictive value for olaparib efficacy (PFS) in first-line ovarian cancer. Methods: The Leuven HRD test was first developed on ovarian cancer tumor samples of the biobank of the University Hospitals Leuven. Then, we analyzed formalin-fixed paraffin-embedded (FFPE) derived DNA from 468 available ovarian cancer samples of the PAOLA-1/ENGOT-ov25 trial. We performed capture-based targeted resequencing of ±90,000 genome-wide single nucleotide polymorphisms (SNPs) at ±40x coverage, and coding exons of BRCA1, BRCA2, RAD51C, RAD51D, PALB2, BLM, BARD1, BRIP1 and TP53 at ±400x coverage. All samples were analyzed using the Leuven HRD and the Myriad test. The BRCAm status, genomic instability score (GIS) and HRD status (comprising both the BRCAm and GIS status) were compared between both tests. The main objective was to compare the predictive value of both tests for predicting PFS in the olaparib versus placebo arm. The Hazard Ratio (HR) and associated 95% confidence interval (CI) were calculated with the use of a Cox proportional hazards model. Conclusions: A robust correlation between the Leuven HRD and Myriad myChoice PLUS test was observed. The Leuven HRD test showed in the PAOLA-1 trial a similar impact on PFS as the Myriad test. Subgroup analyses confirmed the PFS benefit with olaparib in patients with Leuven test BRCAm and HRD-positive/BRCAwt tumors. Objectives: The phase III PAOLA-1/ENGOT-ov25 trial (NCT02477644) evaluated first-line standard therapy including bevacizumab in advanced ovarian cancer with the addition of maintenance olaparib or placebo. Significant improved progression-free survival (PFS) was observed in homologous recombination deficient (HRD) tumors, with or without BRCA mutation (BRCAm), tested with Myriad myChoice PLUS (Myriad test) in contradiction to homologous recombination proficient (HRP) tumors, revealing the need for HRD testing in first line. As part of the ENGOT European HRD initiative, we developed the ‘Leuven’ HRD test. To the best of our knowledge, this is the first presentation of an alternative academic laboratory-developed HRD test compared with the Myriad test including an analysis of the predictive value for olaparib efficacy (PFS) in first-line ovarian cancer. Methods: The Leuven HRD test was first developed on ovarian cancer tumor samples of the biobank of the University Hospitals Leuven. Then, we analyzed formalin-fixed paraffin-embedded (FFPE) derived DNA from 468 available ovarian cancer samples of the PAOLA-1/ENGOT-ov25 trial. We performed capture-based targeted resequencing of ±90,000 genome-wide single nucleotide polymorphisms (SNPs) at ±40x coverage, and coding exons of BRCA1, BRCA2, RAD51C, RAD51D, PALB2, BLM, BARD1, BRIP1 and TP53 at ±400x coverage. All samples were analyzed using the Leuven HRD and the Myriad test. The BRCAm status, genomic instability score (GIS) and HRD status (comprising both the BRCAm and GIS status) were compared between both tests. The main objective was to compare the predictive value of both tests for predicting PFS in the olaparib versus placebo arm. The Hazard Ratio (HR) and associated 95% confidence interval (CI) were calculated with the use of a Cox proportional hazards model. Conclusions: A robust correlation between the Leuven HRD and Myriad myChoice PLUS test was observed. The Leuven HRD test showed in the PAOLA-1 trial a similar impact on PFS as the Myriad test. Subgroup analyses confirmed the PFS benefit with olaparib in patients with Leuven test BRCAm and HRD-positive/BRCAwt tumors.

Introduction/Background Current ESGO guidelines recommend secondary cytoreductive surgery (SCS) followed by chemotherapy in case of first recurrent epithelial ovarian cancer and a platinum-free interval (TFIp) of >6 months as it is the best strategy to prolong progression free survival (PFS) and overall survival (OS). Two prediction models have been developed to improve patient selection for complete resection: AGO and iMODEL. Whole-body diffusion-weighted MRI (WB-DWI/MRI) is a powerful tool to predict resectable disease, however, it has not yet been integrated in the two prediction models. Our aim was to identify the best tool for prediction of resectable disease. Methodology A retrospective cohort study was performed in the University Hospitals Leuven, a tertiary referral centre, using a database search identifying patients between January 2012 and December 2021. Inclusion criteria were: (a) first relapse after 6+ months TFIp, and (b) WB-DWI/MRI. AGO and iMODEL scores were calculated when MRI demonstrated resectable disease. Results In total, 246 patients were included. Based on the WB-DWI/MRI, 124 (50.4%) underwent SCS. The performance of WB-DWI/MRI, AGO, and iMODEL score are summarized in Table 1. WB-DWI/MRI (without the use of any model) had the highest accuracy (89%) compared with the addition of AGO and iMODEL scores: 44.6% (p<0.001) and 80.2% (p=0.54), respectively. Adding the AGO or iMODEL score had a negative effect on both the sensitivity and specificity in predicting resectable disease. Furthermore, when WB-DWI/MRI revealed resectable disease, these patients had a significant longer median PFS: 42.9 months vs. 10.0 months (Hazard Ratio [HR]: 0.35; 95%CI 0.26–0.48) and median OS: 64.9 months vs. 31.4 months (HR: 0.36; 95%CI 0.25–0.53) for resectable versus non-resectable disease, respectively. Conclusion WB-DWI/MRI was the most suitable modality for the prediction of resectable disease at the time of SCS. Adding AGO or iMODEL score did not improve prediction of operable disease in our centre.

Introduction/Background Poly(ADP-ribose)-polymerase inhibitors (PARPi) have changed the treatment landscape for high grade serous ovarian cancer. The CLIO trial (NCT02822157) evaluated olaparib (OLA) single-agent therapy versus physician’s choice chemotherapy (CT) in recurrent epithelial ovarian cancer. Current available tests for homologous recombination deficiency (HRD) have been able to identify possible responders to PARPi, but improvements to these tests are necessary and validation in clinical trials is key. Methodology With Leuven HRD test we provide an academic laboratory-developed method for HRD testing in ovarian cancer. The test was designed on DNA tumor samples of the biobank of University Hospitals Leuven and showed its predictive effect for OLA efficacy in the PAOLA-1/ENGOT-ov25 study (SGO 2022). Here we report the results of Leuven HRD test (LOH+TAI+LST) in the CLIO trial. Results will be compared to Myriad myChoiceDX on the same samples. Results In CLIO 160 patients (60 PSOC and 100 PROC) were randomized 2:1 to OLA (n=107) or CT (n=53). Baseline characteristics were similar between both arms. Overall objective response rate (ORR) for OLA and CT were similar (24.3% and 28.3%, respectively). In PSOC, ORR was 35.0% and 65.0% for OLA and CT (p=0.053); in PROC, ORR was 17.9% and 6.1% for OLA and CT (p=0.134). All patients were tested for germline/somatic BRCA1/2 prior to inclusion. 117 FFPE tumor samples at diagnosis were retrieved and tested for HRD with Leuven HRD test. In PSOC Leuven HRD test was a good predictor of PFS benefit with HR0.35 (p=0.035). There was no difference in PFS in PROC based on Leuven HRD status (p=0.274). Myriad myChoiceDX testing on the same samples is ongoing and comparison of HRD test results will be presented at the meeting. Conclusion Leuven HRD test is predictive for OLA efficacy not only in first-line setting but also in recurrent setting in the CLIO trial.

Introduction/Background Single-agent chemotherapies, like doxorubicin, have very modest activity in recurrent cervical cancer (rCC). Recently, anti-programmed-death protein 1 (anti-PD-1) treatment has shown activity in randomized phase III studies in rCC. In the current study we investigated the combination of doxorubicin with an anti-PD-L1 inhibitor atezolizumab (DA), based on the possible synergistic effect, versus doxorubicin (D) alone. Methodology Prospective open-label, randomized phase II BGOG-cx3 trial (EudraCT2016–000547–14) randomizing 2:1 to doxorubicin (60 mg/m2 q3 wks) with or without atezolizumab (1200 mg q3wks), respectively. The primary endpoint was progression-free survival (PFS) rate at 9 months. Secondary endpoints included objective response rate (ORR), duration of response (DOR), disease control rate (DCR), overall survival (OS), PFS and safety analysis. Results 40 patients were randomized between November 2017 and October 2020: 23 vs 17 patients for DA and D, respectively. Baseline characteristics were similar in both arms (total population: squamous cell carcinoma 84%, prior radiochemotherapy 69%, prior anti-VEGF 61%, median prior lines of chemotherapy in advanced/recurrent setting was 1 with range 0–2). There was a tendency towards a longer median PFS of 4.8 and 3.9 months (figure 1) for DA and D, respectively with HR 0.501 (95%CI 0.246–1.017) (p= 0.0558). Similarly, the primary endpoint, PFS rate at 9 months, was numerically higher but failed to reach significance (26% vs 13% for DA and D, respectively (p=0.054)). Median OS was 10.3 and 7.8 months (p= 0.21) for DA and D, respectively. DCR at 24 weeks was 16% (DA)vs 0% (D) (p=0.279). Results according to PD-L1 staining will be presented. Discontinuation and dose reductions of D were similar in both groups. No new safety signals were noted for the combination of DA. Conclusion Notwithstanding the limited samples size, this study showed a tendency towards a prolonged PFS and OS when doxorubicin was combined with atezolizumab compared with doxorubicin alone in rCC.

Objective. Comparison of olaparib (OLA) monotherapy versus chemotherapy in patients with platinum-sensitive (PSOC) or platinum-resistant ovarian cancer (PROC). Methods. Patients with measurable disease and >= 1 prior line of chemotherapy (CT) were randomized 2:1 to OLA (300 mg tablets, BID) or physician's choice CT.: for PSOC: Carboplatin-Pegylated-Liposomal-Doxorubicin (PLD) or Carboplatin-Gemcitabine; for PROC: PLD, Topotecan, Paclitaxel or Gemcitabine. Results. 160 patients (60 with PSOC and 100 with PROC) were randomized 2:1 to OLA (n=107) or CT (n=53). Baseline characteristics were similar between both arms. Overall objective response rate (ORR) for OLA and CT were similar (24.3% (26/107) and 28.3% (15/53), respectively). Clinical benefit rate (= 12 weeks) was similar with 54.2% (58/107) and 56.6% (30/53), respectively. In PSOC, ORR was 35.0% (14/40) and 65.0% (13/20) for OLA and CT (p=0.053); in PROC, ORR was 17.9% (12/67) and 6.1% (2/33) for OLA and CT (p=0.134). ORR inheavily pretreated PROC (>4 prior lines) was 22.9% (8/35) with OLA versus 0% (0/14) for CT. ORR of 35.7% (5/14) and 13.2% (7/53) was observed in BRCA-mutated and -wildtype PROC cases, respectively. Median PFS in PROC was not significantly different with 2.9 months (95% CI 2.8-5.1 in the OLA group versus 3.8 months (95% CI 3.0-6.4) in the CT group (hazard ratio [HR] 1.11 [95% CI 0.72-1.78]; log-rank p=0.600). Conclusion. OLA monotherapy showed overall an equal response rate in relapsed ovarian cancer compared with CT. In PROC, ORR and TFST tended to be higher with OLA than with CT. In heavily pretreated patients (four lines or more) with PROC disease, OLA treatment seemed to be more effective than CT. (C) 2022 Elsevier Inc. All rights reserved.

Abstract Introduction Accurate information on cause of death is essential for correct breast cancer-specific mortality assessment. However, registration and coding of cause of death is prone to error since determining the exact underlying condition related to the death is challenging. In this study, an expert review of medical files was done to determine the principal cause of death for breast cancer patients of a Belgian tertiary hospital. The retrieved cause of death was compared to death certificate information to assess concordance between both sources. Secondly, the impact of discordant reporting on cause-specific survival (CSS) and other net survival approaches were examined. Methods Breast cancer patients diagnosed and treated at University Hospitals Leuven (UHL) between 2009 and 2014 with follow-up until December 31st, 2016, were included in the study. Information on cause of death was obtained from death certificates (following ICD-10 rules) and medical files. The latter were reviewed by a board of experts at UHL. Agreement was calculated using Cohen’s kappa coefficient, and reasons for discordant reporting were assessed. CSS was calculated based on cause of death information from both sources using the Kaplan-Meier method. These survival estimates were compared to the relative survival probability (RS) using the Ederer II and Pohar Perme method. Results A total of 2,862 patients were included, of whom 354 died after a median follow-up of 54.6 months. We found overall substantial agreement (kappa-value of 0.69 (95% C.I.: 0.62-0.77)) between cause of death reported by death certificates and medical files (Table 1). In 84.8% of cases, there was concordance between both methods. When comparing to medical files, misattribution of breast cancer-specific death in death certificates (4.5% of cases) was linked to the presence of comorbidities (43.7%), metastases (37.5%), or unspecified causes (18.8%). Five-year CSS based on medical files (93.1% (95% C.I.: 91.9-94.1)) was only slightly higher compared to CSS based on death certificates (92.3% (95% C.I.: 91.2-93.4)). RS measures using Ederer II and Pohar Perme were comparable to CSS measures. Conclusions Overall, substantial agreement of cause of death was seen between death certificates and medical files. Attribution of cause death to comorbidities was the most common reason for discordant reporting of breast cancer-specific death. Five-year breast cancer-specific survival was slightly higher based on cause of death information from medical files, compared to death certificates. Periodic reviews and implementation of ICD-10 guidelines for classification of cause of death could improve accuracy in cause of death annotation. Table 1: Discordance for the principal cause of death between medical files and death certificatesmedical filesother causesbreast cancerdeath certificatesother causes136 (38.4%)16 (4.5%)152 (42.9%)breast cancer38 (10.7%)164 (46.3%)202 (57.1%)174 (49.2%)180 (50.9%)354 (100%) Citation Format: Hava Izci, Tim Tambuyzer, Jessica Vandeven, Jérôme Xicluna, Hans Wildiers, Kevin Punie, Nynke Willers, Eva Oldenburger, Els Van Nieuwenhuysen, Patrick Berteloot, Ann Smeets, Ines Nevelsteen, Liesbet Van Eycken, Harlinde De Schutter, Patrick Neven, Geert Silversmit, Freija Verdoodt. Cause of death discordance between death certificates and medical files: Impact on cancer survival assessment in a Belgian case study [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 PS7-63.