Impact of bodyweight (BW)-based starting doses on safety and efficacy of lenvatinib (LEN) in patients (pts) with hepatocellular carcinoma (HCC).

e16119 Background: In Study 202, pts received LEN 12 mg/d for treatment of unresectable (u)HCC, irrespective of BW. Correlations between early dose adjustments and BW in Study 202 led to BW-adjusted dosing in the phase 3 REFLECT study of LEN in uHCC. In REFLECT, pts <60 kg received LEN 8 mg/d and pts ≥60 kg received LEN 12 mg/d. While improvements in safety were expected by reducing the LEN starting dose to 8 mg/d in pts <60 kg, any differences in safety and efficacy of BW-adjusted dosing vs standard dosing among pts with lower BW have yet to be determined. Herein, we compare safety and efficacy of LEN based on pts’ BW (<60 kg vs ≥60 kg) in Study 202 (LEN 12 mg/d irrespective of BW) and in Japanese pts in REFLECT (dose adjusted by BW). Methods: This ad hoc analysis included all pts from Study 202 (n=46; all Asian, 43 Japanese) and Japanese pts from REFLECT (n=81). Safety and efficacy were assessed in Study 202 and REFLECT according to pt BW (<60 kg vs ≥60 kg). Tumors were assessed using mRECIST, by independent radiologic review. Results: In Study 202, pts in the <60 kg group had a median treatment duration of 5.8 mos and received 57% of the LEN planned starting dose (PSD); pts in the ≥60 kg group had a median treatment duration of 9.4 mos and received 78.6% of the LEN PSD. Among Japanese pts in REFLECT, median treatment duration was 5.7 mos in both treatment groups; pts in the LEN 8 mg group received 79.1% of the PSD and pts in the LEN 12 mg group received 70.7% of the PSD. In Study 202, treatment-related adverse events (TRAEs) led to dose reduction in 80.8% of pts in the <60 kg group and 55% of pts in the ≥60 kg group. In REFLECT, TRAEs led to dose reductions in 52.5% of pts in the LEN 8 mg group and 70.7% of pts in the LEN 12 mg group. Serious TRAEs occurred at incidences of 46.2% and 10.0% in Study 202 in the <60 kg and ≥60 kg groups, and 15.0% and 22.0% in REFLECT in the LEN 8 mg and LEN 12 mg groups, respectively. Efficacy results are in the Table. Conclusions: Despite the small sample size, this comparison of Study 202 pts and Japanese pts from REFLECT indicates that BW-adjusted LEN dosing in pts with uHCC yielded comparable efficacy between pts <60 kg who received LEN 8 mg and pts ≥60 kg who received LEN 12 mg. The safety profile was favorable in pts <60 kg who received LEN 8 mg in REFLECT vs those who received LEN 12 mg in Study 202. Fewer serious TRAEs and dose reductions due to TRAEs were observed in pts with lower BW who received the LEN 8 mg starting dose in REFLECT. These data suggest that by reducing the starting dose from LEN 12 mg to 8 mg in pts with uHCC and lower BW (<60 kg), safety is improved without compromising efficacy. Clinical trial information: NCT00946153; NCT01761266. [Table: see text]

Phase II prospective open label study of neoadjuvant pertuzumab, trastuzumab, and nab-paclitaxel in patients with HER-2 positive advanced breast cancer.

583 Background: HER2 overexpression occurs in 20-25% of breast cancers (BC) and is associated with poor prognosis. The addition of trastuzumab (trast) to chemotherapy significantly improves disease-free (DFS) and overall survival (OS) in the adjuvant setting. Pertuzumab (pert) inhibits ligand-activated signaling and in combination with trast has synergistic inhibition of BC cells overexpressing HER2. In the neoadjuvant therapy (NT) setting, the combination of trast, pert, and docetaxel can improve the pCR rate. PCR may predict for improved DFS and OS. De-escalation with weekly paclitaxel combined with trast and pert appeared to be safe and efficacious but requires steroid premedication, whereas nab-paclitaxel (nab) does not require steroid premedication. To decrease treatment-associated toxicity in patients with HER2+ BC, we utilized a non-anthracycline regimen with pert, trast, and nab as NT. The objectives of this study were to evaluate the safety and efficacy of pert added to trast and nab in HER2+ locally advanced BC (LABC) to determine the pCR, as well as DFS and OS. Methods: A total of 45 patients with biopsy-confirmed HER2+ LABC or inflammatory BC were enrolled from 2013-2017, and were treated with 6 cycles of neoadjuvant pert (840 mg loading dose, then 420 mg IV day 1 every 21 days), weekly trast (4 mg/kg loading dose, then 2 mg/kg), and weekly nab (100 mg/m2 IV). Patient characteristics, including age, race, menopausal status, grade, stage, and prior surgery and radiation were recorded. Median treatment cycles determined, and events (AE) were identified for each arm. PCR rate, DFS and OS were calculated. Results: Median age was 56 (31-78) years. 1/45 (2%) was stage I, 30/45 (67%) were stage II, 14/45 (31%) were stage III. pCR rate was 29/45 (64.4%). The initial primary tumor size was similar in pCR and non-pCR patients (mean 4.1 cm vs. 3.2 cm, respectively). Median follow-up was 36.1 months (95% CI [27.1, 41.8]). Median treatment cycles completed was 6 (1-6). A total of 4/45 (9%) patients had >1 cycle delayed, and 32/45 (71%) patients had >1 cycle modified. For the patients achieving pCR, the DFS (95% CI) at 3 years was 85.9% (66.7%, 94.4%) and for those without pCR, it was 87.5% (58.6%, 96.7%). OS was not reached (95% CI [NR, NR]). Grade 3 AEs (> 2 patients) included 7/45 (16%) of patients with hypertension; 4/45 (9%) with anemia; and 2/45 (4%) with diarrhea, ALT, fatigue, or rash. Conclusions: This anthracycline-free regimen which included nab achieved great pCR rate of 64.4% in HER2+ BC patients with fewer treatment-related toxicities. The pCR rate is comparable with docetaxel, carboplatin, trast, and pert (TCHP) therapy in NT setting, but without the treatment-associated toxicities. This suggests we may be able to safely avoid anthracyclines and carboplatin for NT in HER2+ BC patients. The improved pCR did not translate into DFS benefit. Clinical trial information: NCT01730833.

Characterization of long-term responders following treatment with talazoparib (TALA) or physician’s choice of chemotherapy (PCT) in the phase 3 embraca trial.

1029 Background: In the EMBRACA trial (NCT01945775), the poly(ADP-ribose) polymerase inhibitor (PARPi) TALA significantly improved progression-free survival (PFS) versus PCT in patients (pts) with germline BRCA1/2-mutated HER2-negative locally advanced/metastatic breast cancer (BC) (8.6 vs 5.6 months [mo]; hazard ratio [HR, 95% CI] 0.54 [0.41-0.71]; P < 0.0001). Predictive markers for response to PARPi, other than germline BRCA1/2 mutational status, are largely unknown. A previous analysis investigated biomarkers associated with long and short responders in EMBRACA. Here, we report the clinical characteristics of long and short responders. Methods: Pts were randomized 2:1 to TALA or PCT. In this retrospective analysis, pts in the intent-to-treat (ITT) population were mapped into two groups based on response: LONG (pts in TALA arm with overall survival [OS] ≥30 mo and duration of treatment ≥24 mo; pts in PCT arm with OS ≥30 mo); SHORT (pts in either arm with a PFS event [progressive disease by Independent Radiological Facility or death] ≤12 wks). Data cutoff date was Sept 30, 2019. Results: Of 431 pts randomized, 412 pts were treated (286 received TALA; 126 received PCT). In the ITT population, 37 pts receiving TALA and 34 pts receiving PCT were identified as LONG responders; 40 pts receiving TALA and 32 pts receiving PCT were SHORT responders. The Table shows a summary of pt characteristics for LONG and SHORT responders. More pts with HR+ BC and no prior CT for ABC were associated with LONG response; more pts with TNBC and ≥2 prior CT regimens or platinum were associated with SHORT response (Table). Median treatment duration for LONG responders (n = 37, TALA; n = 31, PCT) was 33.5 (24.0-61.4) mo for TALA and 7.6 (1.1-36.3) mo for PCT; 51.4% receiving TALA and 91.2% receiving PCT had subsequent therapy. In SHORT responders, median treatment duration was 2.0 (0.1-5.5) mo (TALA) and 1.4 (0.2-5.6) mo (PCT); 67.5% and 68.8% received subsequent therapy following TALA or PCT, respectively. Conclusions: In this clinical characterization of responders from the EMBRACA study, a higher number of LONG responders had HR+ BC and received no prior CT for ABC. A greater proportion of SHORT responders had TNBC and received ≥2 prior CT regimens or platinum. Further investigation is warranted in a larger number of pts. Clinical trial information: NCT01945775 .[Table: see text]

POS0685 MYCOPHENOLATE MOFETIL IN SYSTEMIC LUPUS ERYTHEMATOSUS PATIENTS: FIVE-YEARS DRUG SURVIVAL IN RENAL AND NON-RENAL INVOLVEMENT

Background:The updated EULAR recommendations for the management of systemic lupus erythematosus (SLE) underline the use of Mycophenolate Mofetil (MMF) in the treatment of different disease related manifestations (1). Several randomized controlled trials have demonstrated the efficacy of MMF in lupus nephritis (LN) patients but only case series and open-labelled trials have analyzed the use of this drug in other than LN features. Moreover, no data are available about the MMF retention rate in a real-life setting.Objectives:The present study aims at evaluating the 5-years drug retention rate (DRR) of MMF in a large monocentric SLE cohort. Secondly, we investigated the influence of MMF in disease activity changes and chronic damage progression.Methods:We performed a longitudinal study including all the SLE patients (ACR 1997 criteria) starting MMF treatment in our Lupus Clinic. Data about indications, mean dosage, duration of treatment and reasons for drug withdrawal were registered. The DRR was estimated using the Kaplan–Meier method. Disease activity and chronic damage were assessed by SLE Disease Activity Index 2000 (SLEDAI-2K) and SLICC Damage Index (SDI), respectively.Results:The present analysis included 162 SLE patients (M/F 22/140, median age at the disease diagnosis 25.5 years, IQR 13). At the beginning of MMF treatment, we registered a median age of 34 months (IQR 21) and a median disease duration of 72 months (IQR 123). The most frequent indications for prescribing MMF were LN (101 patients, 62.3%) and musculoskeletal manifestations (39, 24.1%), followed by neuropsychiatric involvement (10, 6.2%), and others disease related manifestations (12, 7.4%; in particular skin involvement, hematological features, myositis, vasculitis). MMF was administered at a mean daily dosage of 2.1±0.6 grams; no differences in dosage were found between the different indications (p=ns).At the longitudinal analysis, we registered a median treatment duration of 30 months (IQR 55). Figure 1 reported data about DRR: in particular, at 60 months follow-up we observed a DRR of 61.1% for LN patients, which was similar to that registered for patients without renal involvement (NLN) (60.5%; p=ns). Interestingly, the DRR at 60 months was higher in the subgroup of patients treated for joint involvement (75.4%), even without reaching a statistically significant difference. During the observation period, 92 patients (59.2%) discontinued MMF (median treatment duration at discontinuation 25 months, IQR 35). Interestingly, the main cause of withdrawal was the achievement of persistent remission, observed in 20 patients (21.7%), followed by loss of efficacy (19 patients, 20.5%), drug intolerance and pregnancy planning (17 patients for both reasons, 18,4%). Furthermore, our analysis confirmed MMF efficacy, as demonstrated by the significant reduction in SLEDAI-2k values after 4, 12 and 24 months of treatment (p< 0.0001 for all the time-points in comparison with baseline). In addition, MMF resulted able to control chronic damage progression, as demonstrated by the lack of significant increase in SDI values (baseline: 0.6, IQR 1; last observation: 0.93, IQR 1; p=ns).Conclusion:The evaluation of a large SLE cohort demonstrated a good retention rate for MMF. In particular, our results demonstrated that MMF is also a safe and effective drug for SLE manifestation other than LN, in particular for joint involvement. Moreover, it is able to control disease activity and to prevent the progression of chronic damage.References:[1]Fanouriakis A et al. Ann Rheum Dis. 2019 Jun;78(6):736-745.Disclosure of Interests:None declared

Efficacy and safety of 5-fluorouracil (5-FU) / levofolinate / irinotecan (FOLFIRI) for previously treated advanced pancreatic cancer (APC).

407 Background: 5-fluorouracil (5-FU) / levofolinate / irinotecan (FOLFIRI) is one of the preferred regimens for patients (pts) with advanced pancreatic cancer (APC) who received prior gemcitabine-based therapy in the National Comprehensive Cancer Network Guidelines. However, its survival benefit or safety in clinical practice is unclear. Methods: We retrospectively assessed the data of consecutive pts with APC who received FOLFIRI as a second or later-line treatment after gemcitabine-based therapy at Shizuoka Cancer Center between May 2014 and March 2020. Results: The characteristics of 102 pts included in this analysis were as follows: median age (range), 67 (39-78) y; male/female, 55/47; ECOG PS0/1/2, 21/72/9; the number of metastatic sites 0/1/2/3/4, 7/48/35/8/4; unresectable/recurrent, 84/18; UGT1A1 status wild/*6 or*28 heterozygous/homo or double-heterozygous/unknown, 40/40/5/17; treatment line of FOLFIRI 2nd/3rd/4th, 64/32/6. Previous treatment history according to the treatment line of FOLFIRI was as follows: 2nd-line, all patients received GEM-based regimen, GEM plus nanoparticle albumin bound paclitaxel in 63 pts (98.4%) and GEM in 1 (1.6%); 3rd, GEM-based and S1 in 20 (62.5%), GEM-based and 5-FU/levofolinate/oxaliplatin (FOLFOX) in 12 (37.5%); 4th, GEM-based, FOLFOX and S-1 or other agent in 5 (83.3%), 2 GEM-based regimens and S1 in 1 (16.7%). The median treatment cycle was 5 (range 1-55). The median treatment cycle according to the treatment line was as follows: 2nd-line, 7(1-55); 3rd, 4(1-14); 4th, 3.5(1-10). The initial dosage for each cytotoxic agent was as follows: bolus 5-FU injected/omited 72/30; continuous 5-FU 2400/2000/1200 mg/m2, 88/13/1; irinotecan 180/150/120/less than or equal to 100mg/m2, 27/59/13/3. The overall response rate (ORR) and disease control rate (DCR) were 5.9% and 52.9%, respectively. ORR and DCR according to the treatment line were as follows: 2nd-line, 9.3/64.1%; 3rd, 0/68.8%; 4th, 0/50.0%. At the median follow up 6.5 M, the median overall survival (OS) and progression free survival (PFS) were 6.6M and 3.1M, respectively. The median OS and PFS according to the treatment line were as follows: 2nd-line, 8.1/3.6M; 3rd, 5.1/2.1M; 4th, 6.6/2.0M. Adverse events (AEs) were observed in 70.8% pts. Grade 3 or higher AEs occurred in 27.2% pts [neutropenia in 26 (25.2%) pts, febrile neutropenia in 4 (3.9%) pts, nausea in 4 (3.9%) pts, decreased appetite in 3 (2.9%) pts]. No treatment related deaths were observed. Conclusions: FOLFIRI is well tolerated and effective especially in the second-line treatment for pts with gemcitabine-refractory APC.

Management of Patients with Relapsed/Refractory Multiple Myeloma (RRMM) Treated with Novel Therapies in Routine Clinical Practice in Germany

Background: To better understand the clinical practice in Germany with respect to the rapidly evolving treatment landscape, we aim to describe the characteristics and treatment patterns of patients (pts) with relapsed/refractory multiple myeloma (RRMM) in Germany. Methods: In a national retrospective medical chart review including consecutive pts treated for RRMM in participating hospitals/centers and practices across Germany between Oct 2017 and Jun 2018, the following data were extracted from Oct 2019 to Feb 2020: pt demographics, disease characteristics, treatment history at MM diagnosis and at initiation of RRMM therapy, and distance to care. Physician's assessment of response to therapy and minimal residual disease (MRD) testing were also collected. Because this analysis was not powered to compare between regimens or lines of therapies, results are purely descriptive. Results: Physicians from 47 participating centers extracted 484 pt charts, that included 214 pts (44%) in 2nd line of therapy (2L), 160 (33%) in 3rd line (3L) and 110 (23%) in 4th line and beyond (4L+) (Figure 1). Half of the participating centers were office-based (52%) and 40% were hospitals. Results are summarized in Table 1 and regimens described by line of therapy in Figure 2: Among the 2L subgroup (N=214), median age was 72 years and the majority (73%) of pts had an ECOG status of 0-1 at 2L initiation. About one-third (31%) of pts had had a transplant. Carfilzomib (CFZ) or daratumumab (Dara)-based regimens were mainly used in 2L pts (40% and 41%, respectively). Among the 87 Dara pts, the triplet regimen Dara, lenalidomide (LEN) and dexamethasone (DRd) was used in more pts (58%) than the triplet with Dara, bortezomib (BTZ) and dexamethasone (DVd, 15%). Among the 85 CFZ pts, nearly half (47%) received the triplet regimen CFZ, LEN and dexamethasone (KRd) and 38% of CFZ pts received the doublet of CFZ and dexamethasone (Kd). Among the remaining 42 pts in 2L, other LEN use was described in 35 pts and BTZ in 7 pts. For 107 pts who completed 2L, median treatment duration was 8 months (9 months for CFZ pts; 8.5 months for Dara pts; 6 months for LEN pts). Where response was available (n=191/214), a complete response (CR) or very good partial response (VGPR) was achieved by 64% of pts, including 73% of Dara pts (n=53/73), 62% of CFZ pts (n=48/78), and 53% of LEN pts (n=18/34). 149 ptssteoprotective drugs at 2L initiation, mainly intravenous zoledronate (ZA, 44%) and subcuteaneous denosumab (Dmab, 42%). In the 3L subgroup (N=160), median age was 72 years. Over half (58%) of pts had an ECOG status of 0-1 at 3L initiation. Most 3L pts received Dara (51%, n=81) or CFZ (35%, n=57), and a further 9% (n=15) of 3L pts received other LEN combinations. In 3L, CFZ was more often used as a doublet (54% Kd vs. 28% KRd); Dara was mainly given in triplet regimens (23% DRd, 17% DVd vs. 12% Dara mono). For 118 pts who had completed 3L at time of data extraction, median treatment duration was 10 months (10 months for both CFZ and Dara pts, 7 months for LEN pts). Where response was available (n=139), CR/VGPR was achieved by 55% of pts in 3L, including 64% of Dara pts (n=46/72) and 41% of CFZ pts (n=20/49). Osteoprotective drugs (51% Dmab, 36% ZA) at 3L initiation were given in 69% (n=111) of pts. In the 4L+ subgroup (N=110), median age was 74 years. A high percentage (64%) of pts had an ECOG status ≥2 at 4L initiation. Over half (52%, n=57) received Dara, mostly as monotherapy (70% Dara; 9% DRd, 2% DVd), followed by CFZ in 26% (n=28) and pomalidomide (POM) in 16% (n=18) of pts in 4L+. For 97 pts who completed 4L at time of data extraction, median treatment duration was 8 months (9 months for both CFZ and Dara pts, 7 months for POM pts). Where response was available (n=106), 39% of pts achieved a CR/VGPR, including 46% of Dara pts (n=25/55) and 36% of CFZ pts (n=10/28). At 4L initiation, 66% (n=72) of pts were receiving osteoprotective drugs (51% Dmab, 40% ZA). Conclusion: This chart review indicates that pts with RRMM who underwent a MM therapy between Oct 2017 and Jun 2018 in routine practice across Germany, mostly received combination regimens with novel agents irrespective of the line of therapy. Lenalidomide is still often combined with novel agents. Patterns of treatment regimens differed by line of therapy and are adapted as disease progresses and ECOG status increases. Disclosures Steinmetz: Amgen; BMS, Celgene, Novartis; Janssen-Cilag; Omnicare, Vifor: Speakers Bureau; Amgen: Membership on an entity's Board of Directors or advisory committees; Alexion, Amgen, Bayer, BMS, Boehringer, Celgene, Janssen-Cilag, Novartis, Omnicare, Sanofi, Takeda: Other: TRAVEL, ACCOMMODATIONS, EXPENSES; DGHO, ESMO, DGPM, BNHO, NIONo: Other: Memberships; Alexion, Accord Healthcare, Amgen; Ariad, Bluebird Bio, BMS, Boehringer, Celgene, Hexal-Sandoz, Novartis; Janssen-Cilag; Omnicare, Oncopeptides, Otsuka, Pfizer, Sanofi, Shire, TAD, Takeda, Vifor: Consultancy; Accord Healthcare, Amgen, Celgene, Novartis, Vifor: Research Funding. Singh:Amgen Ltd, UK: Current Employment. Milce:Amgen: Research Funding; Kantar Health, France: Current Employment. Haidar:Kantar Health, France: Current Employment; Amgen: Research Funding. Rieth:Amgen GmbH, Germany: Current Employment; Amgen: Current equity holder in private company. Lebioda:Amgen GmbH, Germany: Current Employment; Amgen: Current equity holder in private company.

278. Comparison of Oral Cephalexin Versus Oral Levofloxacin for Treatment of Uncomplicated Enterobacterales Bacteremia

Background Aerobic gram-negative rods (GNR) are a common cause of bloodstream infections. While the use of oral antibiotics as step-down therapy for GNR bacteremia is generally accepted, there is debate as to which agents are best. Fluoroquinolones have high oral bioavailability but also have concerns for toxicity, while oral beta-lactams have lower achievable blood concentrations and require more frequent dosing. At Kaweah Delta Medical Center, patients are transitioned from IV antibiotics to oral levofloxacin or cephalexin. The purpose of this study is to determine if these two oral agents are comparable when used as step-down therapy for uncomplicated GNR bacteremia. Methods This was a single-center, retrospective study of adult patients who were admitted from May 2018 - May 2019 with a positive GNR blood culture. The primary objective was to assess treatment failure defined as a composite outcome of readmission due to recurrent bacteremia or mortality within 60-days. Secondary outcomes included subsequent Clostridioides difficile infection, emergence of resistant GNR bacteria, and hospital length of stay. Results A total of 82 patients met inclusion criteria. The combined median age of both treatment groups was 58.2 years (interquartile range [IQR], 46.7–73.5) and greater than 60% of patients were female. Most patients were Hispanic (45.1%) or non-Hispanic white (43.9%). The most common documented source of GNR bacteremia was the urinary tract (74.4%), and the most frequently recovered organism was Escherichia coli (78.0%). Patients were treated with IV antibiotics for a median treatment duration of 4 days (IQR, 4–5) and oral antibiotics for a median treatment duration of 10 days (IQR, 7.3–11) and 9 days (IQR, 7–11) for cephalexin and levofloxacin, respectively. The composite outcome occurred in 2 patients (4.3%) from the cephalexin group and 0 patients from the levofloxacin group (p=0.50). Positive cultures for resistant GNR were found for 3 patients in the cephalexin group. No patients developed subsequent C. difficile infections. Conclusion Patients who received cephalexin or levofloxacin did not have a significant difference in the composite primary outcome. These findings suggest that oral cephalexin is an effective step-down option to treat uncomplicated GNR bacteremia. Disclosures All Authors: No reported disclosures

Safety and efficacy of BAY1436032 in IDH1-mutant AML: phase I study results

The mutant IDH1 (mIDH1) inhibitor BAY1436032 demonstrated robust activity in preclinical AML models, supporting clinical evaluation. In the current dose-escalation study, BAY1436032 was orally administered to 27 mIDH1 AML subjects across 4 doses ranging from 300 to 1500 mg twice-daily. BAY1436032 exhibited a relatively short half-life and apparent non-linear pharmacokinetics after continuous dosing. Most subjects experienced only partial target inhibition as indicated by plasma R-2HG levels. BAY1436032 was safe and a maximum tolerated dose was not identified. The median treatment duration for all subjects was 3.0 months (0.49–8.5). The overall response rate was 15% (4/27; 1 CRp, 1 PR, 2 MLFS), with responding subjects experiencing a median treatment duration of 6.0 months (3.9–8.5) and robust R-2HG decreases. Thirty percent (8/27) achieved SD, with a median treatment duration of 5.5 months (3.1–7.0). Degree of R-2HG inhibition and clinical benefit did not correlate with dose. Although BAY1436032 was safe and modestly effective as monotherapy, the low overall response rate and incomplete target inhibition achieved at even the highest dose tested do not support further clinical development of this investigational agent in AML.