MIRV is a first-in-class antibody-drug conjugate comprising an FRα-binding antibody, cleavable linker, and maytansinoid DM4 payload for patients (pts) with FRα-positive, high-grade serous ovarian cancer. An ER analysis was conducted across 3 studies to understand the impact of MIRV, DM4, and S-methyl-DM4 (SmDM4) metabolite exposure on efficacy and safety in pts with FRα-positive tumors. A validated population pharmacokinetic model was used to derive the exposure metrics including area under the concentration-time curve (AUC) and trough concentration for eligible pts in 3 studies, 2 studies (n=445) with extensive sampling and a 3rd study (n=97) with sparse sampling. The objective response rate (ORR) and progression free survival (PFS) were analyzed as well as adverse events (AE) of grade 2 or higher. Logistic regression and Cox proportional hazard approach were used, and covariate search was performed. The AUC and trough concentration of MIRV were found to correlate with efficacy. Higher exposure to MIRV resulted in higher probability of therapeutic response in both study pools in terms of ORR (p<0.01) and PFS (p<0.03). In the safety analysis, MIRV AUC was the exposure parameter that correlated best with ocular AEs. The incident rate of grade 2+ ocular AEs increased with increasing MIRV exposure (p<0.01 for the pooled analysis). The overall ER relationship for peripheral neuropathy was flat. MIRV exposure did not have a significant impact on pneumonitis. The exposures of the payload, DM4, or the metabolite, SmDM4, were not found to associate with either efficacy or safety. Covariates, including patient demographics and clinical laboratory test results, identified in this model had limited impact of efficacy and safety, and are likely not of clinical significance. ER analysis demonstrated that both efficacy and probability of ocular toxicity correlate with exposure of MIRV. The analyses support the final recommended dose of 6 mg/kg based on adjusted ideal body weight every 3 weeks with balanced efficacy and safety.

MIRV is a first-in-class antibody-drug conjugate comprising an FRα-binding antibody, cleavable linker, and maytansinoid DM4 payload for patients (pts) with FRα-positive, high grade serious ovarian cancer (OC). Analysis was conducted across 3 studies to understand the impact of pt characteristics on the PK parameters of MIRV, DM4, and S-methyl-DM4 (SmDM4) metabolite in pts with FRα-positive tumors. A base model was developed for MIRV followed by a stepwise covariate search. Structural components were added for DM4 and SmDM4 concentration data while fixing MIRV PK parameters. The final semi-mechanistic model accommodated saturable elimination of MIRV and the payload metabolite. Model construction was based on 2 studies (n=445) with extensive sampling and validated in a 3rd study (n=98) with sparse sampling. The evaluated doses ranged from 0.15 mg/kg to 7 mg/kg total body weight, including the final recommended dose of 6 mg/kg adjusted ideal body weight (AIBW) every 3 weeks, which was administered to 87.8% of total pts. The population PK model predicted a low clearance (CL), small volume of distribution (Vd), and long elimination half-life (t1/2) for MIRV. Statistically significant covariates include AIBW, serum albumin, and age. Tumor FRα expression was not a significant covariate. AIBW showed an impact on CL and Vd. Albumin showed significant effect on CL but not on Vd. Age did not impact CL and was found to be a minor covariate on Vd. For 6 mg/kg AIBW, the effects of these covariates on the exposures to MIRV do not require dose adjustment. With a terminal elimination t1/2 of 115 hrs (4.8 days), the steady state can be reached in ∼24 days. Mild or moderate renal impairment and mild hepatic impairment did not significantly alter PK. Coadministration with weak, moderate, and strong cytochrome P450 3A4 (CYP3A4) inhibitors did not have clinically meaningful effects on PK parameters for MIRV, DM4, or SmDM4; CYP3A4 inducers were not evaluated. Population PK analysis supports 6 mg/kg AIBW dosing of MIRV, with exposure to MIRV, DM4, and SmDM4 maintained within the target range. Dosing adjustments would not be required for mild or moderate renal or mild hepatic impaired population.

Biomarkers of cellular proliferation (eg. Ki67≥20%) may be useful in estimating recurrence risk in HR+, HER2- early breast cancer (EBC: Stage I-III). The use of chemotherapy (CT) depends on multiple factors including the patient’s (pt) estimated risk of recurrence. One area of uncertainty in risk estimation is in pts with 1-3 positive nodes. This study aimed to understand the use of Ki67 testing and scoring in pts with node+, HR+, HER2- EBC and treatment (trt) patterns. An observational retrospective cohort study of HR+, HER2- EBC pts treated (JAN2010-AUG2018) in community practices utilized molecular and clinical data aggregated via Genospace, a web-based data analysis platform. Inclusion criteria: pts (>18 yrs) diagnosed with HR+, HER2- invasive EBC. Demographic, pathologic, and trt data were collected. Multivariable logistic regression (MLR) was used to analyze predictors of testing and Ki67 scores≥20%. 555 (98%) of 567 pts (mean age 58y, SD 12.7y) with initial tumor stage I (63), II (370), III (121), unknown (13) received systemic therapy for EBC. 91 (16%) pts received neo-adj therapy. Adj therapy was received by 525 (95%) pts, with 428 (82%) receiving endocrine therapy (ET) in the regimen and 175 (33%) treated with CT followed by ET. Among pts with 1–3 positive nodes (n=258), 24% (61) were tested for Ki67 expression. Of these, 54% (33/61) had scores of ≥20%. Following Ki67 testing, 59% pts had CT and 67% pts with Ki67≥20% received CT. MLR analyses showed that clinical or pathological factors were not predictors of Ki67 testing. Individuals without insurance were less likely to receive Ki67 testing (HR -2.7, p<0.0003). Of those tested (n=130), grade 2 (p=0.0027) and grade 3 (p<0.001) were predictors of Ki67 scores ≥20%. Overall, Ki67≥20% was common among those with grade 2 (64%) or grade 3 (94%) tumors, however, 28% of pts with grade 1 tumors also had Ki67≥20%. HR+ EBCs are routinely treated with ET+/-CT. Markers of cellular proliferation, such as Ki67, are prognostic of risk of recurrence. Clinical and pathological factors were not predictors of Ki67 testing. Higher Ki67 scores were not limited to high grade tumors. The role of high Ki67 in general, and in lower grade tumors, remains unclear.

Among patients [pts] with early-stage HR+, HER2- breast cancer [BC], this study aimed to measure associations of clinical characteristics, pts' perception of treatment [tx] goals and endocrine therapy [ET] history on pts' health-related quality of life [HRQoL]. A multinational (France, Germany, Italy, Spain, UK, Japan and US) survey of pts diagnosed [dx] with stage I-III HR+, HER2- BC was conducted from June to October 2019. Pts identified by their physician were invited to complete a pen and paper questionnaire that included the FACT-G and questions on awareness of tx goals. Four separate multivariable linear regression analysis were conducted to measure the association between patients' clinical characteristics and ET history on each of the 4 FACT-G domain scores: Physical Wellbeing [PWB], Social/Family Wellbeing [SWB], Emotional Wellbeing [EWB] and Functional Wellbeing [FWB]. Regression coefficients are reported where p values <0.05. A positive coefficient implies higher HRQoL. 1152 pts were recruited by 320 physicians. PWB (n=1030) score was impacted if pts completed ET ≤6 months and >6 months before data collection (+2.87, +3.16), if BC dx Stage IIIA or IIIB/C (-2.32 and -2.38) and if pts felt goal of their tx was to stop BC from progressing (-3.00). SWB (n=1029) score was impacted if pt currently receiving tamoxifen (+2.02), if BC dx Stage IIIB/C (-2.69) and if pts felt goal of their tx was to prevent BC returning after surgery (+1.19). EWB (n=1033) score was impacted by time since dx (+0.001), if BC dx Stage IIIB/C (-1.87), or if pts felt goal of their tx was to stop BC from progressing (-2.94). FWB (n=1034) score was impacted if pts were older (-0.10), by time since dx (+0.002), if pre/peri-menopausal (-1.76), if BC dx Stage IIA, IIIA or IIIB/C (-2.18, -3.09, -4.07), if pts felt goal of their tx was to stop BC from progressing (-2.65) or increase length of life (-1.16). The 4 FACT-G HRQoL domain scores were impacted by stage of BC at diagnosis, ET history and pt perception of tx goals. This suggests that earlier diagnosis and positive reinforcement of tx goals could have a positive impact on HRQoL.

For patients with early breast cancer (EBC), genomic testing is increasingly informing the likelihood of benefit from adjuvant (adj) chemotherapy (CT). The objective was to examine, by geographic region, genomic/biomarker testing rates and treatment patterns in HR+, HER2- EBC patients in the adj setting. Real-world data from France, Germany, Italy, Spain, United Kingdom (EU5), Japan, and United States (US) were drawn from the Adelphi EBC I Disease Specific Programme, a point in time observational study. 320 physicians completed patient record forms (Jul to Oct 2019) for up to the next 8 eligible HR+, HER2- EBC patients receiving or had received adj therapy in the last 12 months, prior to data collection. Data included demographics, genetic/biomarker testing, and treatment received. N = 2447 patients (98% female, mean age 59.6 yrs, SD 13.0). The frequency of neo-adjuvant (n-adj) therapy varied by region. Adj treatment patterns differed by region and by the use of n-adj therapy. Breast cancer gene (BRCA) testing was more frequent in patients who received n-adj therapy and more common in the US than EU5 and Japan. EU5 had the highest Ki67 testing rates. OncotypeDx was the most frequent multi-gene assay, more so in the US. Mammaprint was less frequent and Adjuvant! Online more frequent in the EU5 than the US and Japan (Table).Table: 174PTest*EU5 n=1859US n=347Japan n=241N-adj** n=293 (15.8%)Adj only n=1566 (84.2%)N-adj** n=92 (26.5%)Adj only n=255 (73.5%)N-adj** n=44 (18.3%)Adj only n=197 (81.7%)BRCA1100 (34.1%)236 (15.1%)42 (45.7%)96 (37.6%)8 (18.2%)13 (6.6%)BRCA291 (31.1%)209 (13.3%)31 (33.7%)87 (34.1%)2 (4.5%)3 (1.5%)Ki67195 (66.6%)1121 (71.6%)31 (33.7%)73 (28.6%)9 (20.5%)85 (43.1%)OncotypeDx57 (19.5%)223 (14.2%)31 (33.7%)125 (49.0%)13 (29.5%)30 (15.2%)MammaPrint16 (5.5%)74 (4.7%)10 (10.9%)15 (5.9%)10 (22.7%)9 (4.6%)Adjuvant! Online23 (7.8%)120 (7.7%)2 (2.2%)7 (2.7%)1 (2.3%)26 (13.2%)*EndoPredict, Breast Cancer Index, Mammostrat, PAM50 omitted (% use <10%)**Excludes non-Endocrine Therapy Open table in a new tab *EndoPredict, Breast Cancer Index, Mammostrat, PAM50 omitted (% use <10%) **Excludes non-Endocrine Therapy The use of genomic/biomarker testing varied by region. Variation between regions in treatment approach may be driven, in part, by genomic testing/biomarker results as testing is increasingly being used to minimize over and under treatment with CT. Treatment patterns and other factors predicting treatment will be analyzed.

Among patients [pts] diagnosed with early-stage HR+, HER2- breast cancer (BC), this study aimed to describe awareness, information sources, satisfaction and degree of involvement in treatment decisions. A multinational (France, Germany, Italy, Spain, UK, Japan and US) survey of pts diagnosed with stage I-III HR+, HER2- BC was conducted from June to October 2019. Pts identified by their physician were invited to complete a pen and paper questionnaire with questions on awareness of 4 aspects of their BC (stage, nodal status, HER2 and HR status), information sources used, satisfaction, and involvement in treatment decisions for their BC. 1152 pts completed the questionnaire (mean age 59 years; 33% with degreed education; 31% pre/peri-menopausal and 1% male; Stage at diagnosis: I 30%, II 48%, III 22%). UK pts were most aware of all 4 aspects of their BC (58%), Italian (18%) and Japanese (20%) pts the least aware. Pts reporting being actively involved in treatment decisions was highest in Germany (68%) and least in Spain (30%). Awareness of all 4 aspects of their BC was highest among pts with degreed education (48%), with 63% reporting active involvement in treatment decisions; and lowest amongst those with no degreed education (30%), with 47% reporting active involvement. Pre/peri-menopausal pts were most aware of all 4 aspects of their BC (46%) (post-menopausal pts (31%). 90% of all pts cited their doctor as an information source. The second most common source was the internet amongst pre/peri-menopausal pts (57%) and family/friends for post-menopausal pt (39%). 71% who very much agree with the statement 'I am satisfied with how well I am coping with my illness' feel they are actively involved in treatment decisions compared to 44% of who do not agree at all with the statement. A high proportion of pts satisfied with how they were coping with their illness were actively involved in their treatment decisions. Knowledge and involvement were highest in pts with degreed education and amongst pre/peri-menopausal pts, suggesting there may be a need to raise knowledge and awareness in older pts or those without a degreed education.

3024 Background: Chemotherapy (C) induced bone marrow suppression has been assumed to inhibit the generation of augmented immune responses. This has limited investigation of front-line C-IT in OC. Temporal relationship of IT administration relative to C has not been established. Oregovomab is a CA125 specific MAb in development as an OC adjuvant treatment to induce tumor specific immunity. Methods: Patients (pts) awaiting staging laparotomy for presumptive OC were consented to the study and contributed primary tumor samples. Qualifying pts were subsequently randomized to concurrent C-IT with antibody administered at cycles 1, 3, and 5 (Arm A) or 8 ± 2 days following cycles 1, 3, and 5 (Arm B) of standard C-P. Both arms received additional infusions of oregovomab every 12 weeks until progression; response to C-IT was assessed at 36 weeks. Results: Forty stage III/IV pts (18 Arm A; 22 Arm B) were randomized and dosed. More pts in Arm A (39%) were stage IV vs Arm B (14%) however 27% of Arm B pts had >2cm residual disease vs. 11% in Arm A. Complete clinical response (CR) to surgery and C-IT was achieved in 15 pts (83%) in Arm A and 17 pts (77%) in Arm B. For pts with = 2 cm residual disease, 88% in Arm A had = 65 U/mL CA125 prior to/at cycle 3, normalized CA125 and a CR post C-IT compared to 75% of pts in Arm B. Robust Ab2 antibody response (>100 ng/mL) was achieved in 94% of Arm A and 68% of Arm B pts. The immune response developed earlier in Arm A. After one injection of oregovomab, 31% of pts in Arm A vs. 0% in Arm B generated HAMA > 3000 ng/mL and Ab2 > 100 ng/mL (P=0.016 Fisher’s Exact Test). Such an early and vigorous response pattern has not been seen in previous oregovomab studies which have not included concomitant front-line C. There were no IT related serious adverse events. Conclusions: Contrary to expectations, concurrent C-P resulted in enhanced immune stimulation with oregovomab. Interestingly, concurrent infusion of oregovomab was more immunogenic than delayed infusion. Clinical activity was favorable and safety profiles were similar to that expected for C alone. Further randomized evaluation of front-line CP-oregovomab vs CP is warranted. [Table: see text]