Search

PURPOSE Rogaratinib, an oral pan-fibroblast growth factor receptor (FGFR1-4) inhibitor, showed promising phase I efficacy and safety in patients with advanced urothelial carcinoma (UC) with FGFR1-3 mRNA overexpression. We assessed rogaratinib efficacy and safety versus chemotherapy in patients with FGFR mRNA-positive advanced/metastatic UC previously treated with platinum chemotherapy. METHODS FORT-1 (ClinicalTrials.gov identifier: NCT03410693 ) was a phase II/III, randomized, open-label trial. Patients with FGFR1/ 3 mRNA-positive locally advanced or metastatic UC with ≥ 1 prior platinum-containing regimen were randomly assigned (1:1) to rogaratinib (800 mg orally twice daily, 3-week cycles; n = 87) or chemotherapy (docetaxel 75 mg/m2, paclitaxel 175 mg/m2, or vinflunine 320 mg/m2 intravenously once every 3 weeks; n = 88). The primary end point was overall survival, with objective response rate (ORR) analysis planned following phase II accrual. Because of comparable efficacy between treatments, enrollment was stopped before progression to phase III; a full interim analysis of phase II was completed. RESULTS ORRs were 20.7% (rogaratinib, 18/87; 95% CI, 12.7 to 30.7) and 19.3% (chemotherapy, 17/88; 95% CI, 11.7 to 29.1). Median overall survival was 8.3 months (95% CI, 6.5 to not estimable) and 9.8 months (95% CI, 6.8 to not estimable; hazard ratio, 1.11; 95% CI, 0.71 to 1.72; P = .67). Grade 3/4 events occurred in 37 (43.0%)/4 (4.7%) patients and 32 (39.0%)/15 (18.3%), respectively. No rogaratinib-related deaths occurred. Exploratory analysis of patients with FGFR3 DNA alterations showed ORRs of 52.4% (11/21; 95% CI, 29.8 to 74.3) for rogaratinib and 26.7% (4/15; 95% CI, 7.8 to 55.1) for chemotherapy. CONCLUSION To our knowledge, these are the first data to compare FGFR-directed therapy with chemotherapy in patients with FGFR-altered UC, showing comparable efficacy and manageable safety. Exploratory testing suggested FGFR3 DNA alterations in association with FGFR1/ 3 mRNA overexpression may be better predictors of rogaratinib response.

PURPOSE Antitumor activity in preclinical models and a phase I study of patients with dedifferentiated liposarcoma (DD-LPS) was observed with selinexor. We evaluated the clinical benefit of selinexor in patients with previously treated DD-LPS whose sarcoma progressed on approved agents. METHODS SEAL was a phase II-III, multicenter, randomized, double-blind, placebo-controlled study. Patients age 12 years or older with advanced DD-LPS who had received two-five lines of therapy were randomly assigned (2:1) to selinexor (60 mg) or placebo twice weekly in 6-week cycles (crossover permitted). The primary end point was progression-free survival (PFS). Patients who received at least one dose of study treatment were included for safety analysis (ClinicalTrials.gov identifier: NCT02606461 ). RESULTS Two hundred eighty-five patients were enrolled (selinexor, n = 188; placebo, n = 97). PFS was significantly longer with selinexor versus placebo: hazard ratio (HR) 0.70 (95% CI, 0.52 to 0.95; one-sided P = .011; medians 2.8 v 2.1 months), as was time to next treatment: HR 0.50 (95% CI, 0.37 to 0.66; one-sided P < .0001; medians 5.8 v 3.2 months). With crossover, no difference was observed in overall survival. The most common treatment-emergent adverse events of any grade versus grade 3 or 4 with selinexor were nausea (151 [80.7%] v 11 [5.9]), decreased appetite (113 [60.4%] v 14 [7.5%]), and fatigue (96 [51.3%] v 12 [6.4%]). Four (2.1%) and three (3.1%) patients died in the selinexor and placebo arms, respectively. Exploratory RNA sequencing analysis identified that the absence of CALB1 expression was associated with longer PFS with selinexor compared with placebo (median 6.9 v 2.2 months; HR, 0.19; P = .001). CONCLUSION Patients with advanced, refractory DD-LPS showed improved PFS and time to next treatment with selinexor compared with placebo. Supportive care and dose reductions mitigated side effects of selinexor. Prospective validation of CALB1 expression as a predictive biomarker for selinexor in DD-LPS is warranted.

5572 Background: AcSé Pembrolizumab is a Phase 2, non-randomized parallel arms, multicentric basket trial investigating the efficacy and safety of pembrolizumab monotherapy in different cohorts of patients with rare cancers (NCT03012620). Here we report the results in the rare ovarian tumors cohort. Methods: Selected histotypes were all rare ovarian cancers (incidence < 6/100,000/year). Main inclusion criteria were age > 18, ECOG PS≤1, resistant disease to platinum based chemotherapy, and systematic histological central review by expert pathologist from TMRG network. Patients (pts) received pembrolizumab 200 mg IV on Day 1 of every 21-day cycle for a maximum of 2 years. The primary endpoint was the confirmed objective response rate according to RECIST v1.1 at 12 weeks. Secondary endpoints included best response rate, duration of response, progression-free survival (PFS), overall survival (OS), and safety. The 7 subgroups of pts analyzed were carcinosarcoma (CS), clear cell carcinoma (CCC), low grade serous carcinoma (LGSC), mucinous carcinoma (MEOC), sex cord tumors (SCT), germ cell tumor (GCT), and smarcA4 deficient hypercalcemic ovarian tumor (SCHOCCT). Results: 62 pts from 22 centers, were included from 08/2017 to 12/2020. Median Age was 53.5 years old [36-64]. Median number of previous lines of chemotherapy was 2 (range 1-4). The median number of cycles was 8 (range, 1-35) with 44 pts (70.9%) who discontinued the trial after a mean number of 6.8 cycles. There were 2 pts (3.2%) with partial response (PR) at 12 weeks. The best response in ITT was complete response (CR) in 1 patient (1%), PR in 3 (14.3%), and stable disease (SD) in 21 (33.8%). The occurrence of best response depended on the histotype with 1 CR (33%) in GCT (cancerized teratoma), 2 PR (20%) in CCC, and 1 PR (4%) in LGSC. 4/4 pts (100%) reported PD as best response in SCOOHT (Table 1). Median duration of response or stabilized disease was 7.8 months [IQR, 4.1 to 9.0]. At the data cut off, 6-month PFS was 29% [19.7-42.8] and 6-month OS was 77.8% [67.7-89.3] on the overall population. Outcomes differed according to subgroups and will be presented. There were a total of 62 adverse events (AEs) reported in 28 pts. For 5 pts (8%) AEs lead to drug discontinuation. AEs were of grade 1 (n = 9), grade 2 (n = 8), or grade ≥ 3 (n = 45: 42 grade 3, 2 grade 4, and 1 grade 5). Conclusions: Pembrolizumab is safe and well tolerate in this population of rare ovarian cancer pts. AcSé study reports prolonged responses in very selected subtypes of rare ovarian tumor (CCC, cancerized teratoma, and LGCS). Acknowledgements: TMRG (national cancer network dedicated to rare gynecological tumors), GINECO group for partnership, La Ligue Nationale contre le Cancer, INCa and MSD. Clinical trial information: NCT03012620. [Table: see text]

3130 Background: PROFILER-02 is a multicenter randomized prospective study comparing the proportion of metastatic cancer patients (pts) with Targeted Agent (TA) recommendation provided by large NGS panel (FOne panel, 324 genes) vs home 87-gene NGS panel (CTRL) (PMID 30865223). Methods: Adult pts with advanced/metastatic cancer during 1st or 2nd line of therapy without known targetable gene alteration were eligible and randomized (1:1) to FOne vs CTRL panel. Both panels were performed for each patient. The randomization arm defined the first panel reviewed by dedicated Molecular Tumor Board (MTB) at disease progression while the 2nd panel remained blinded. The primary objective was the pts rate with at least one TA recommendation by the MTB using either FOne or CTRL panel. The study was designed in order to detect difference in proportions of 10% between the two panels. A sample size of 289 pts with both panels were requested to show this difference with an expected proportion of discordant pairs of 20% using a McNemar's test with 98% power and 5% two-sided significance level. Secondary endpoints included number of pts receiving at least one TA, progression free survival (PFS) and overall survival (OS). Results: From June 2017 to June 2019, among the 339 included pts 171 and 168 pts were randomized in FOne or CTRL panels’ first use, respectively. Median age was 57 years [19.0 - 85.0]; 54.9% were female. The median time from randomization to first MTB was 7.62 months [range 0.80 - 48.1]. Among the 339 pts, 147 pts (43.4%) had no TA recommendation, 108 pts (31.9%) had at least one TA recommendation according to both panels, 67 pts (19.8%) had one or more TA recommendation according to FOne panel only and 17 pts (5%) according to CTRL panel only (McNemar p < 0.001). At the time of the analysis, 51/339 (15%) pts started recommended treatment: 27 pts (8%) with TA recommendation from both panels, 21 pts (6.2%) from FOne only and 3 pts (0.3%) from CTRL only. Main initiated TA were PARP inh. (FOne n = 12; CTRL n = 9), PI3K/AKT/mTOR inh. (FOne n = 10; CTRL n = 9) and immunotherapy (ICI) (FOne n = 7; CTRL n = 0). Median PFS following first MTB were 3.2 months (95% CI 2.5-3.8) and 2.6 months (95% CI 2.0-3.8), median OS were 8.7 months (95% CI 6.6-10.8) and 8.4 months (95% CI 6.4-9.7), in the FOne and CTRL arm, respectively. Conclusions: Larger NGS panel including Tumor Mutational Burden increased the number of recommended options (TA and ICI), as well as the number of treatment initiation. Clinical trial information: NCT03163732.

TPS11585 Background: Primary metastatic osteosarcoma (OS) patients are treated with a curative intent following the same principles of non-metastatic OS, while the treatment of recurrent OS is primarily surgical in the case of isolated lung metastases. When complete removal of all metastases cannot be achieved, the prognosis remains poor, with a median Progression-Free Survival (PFS) between 3 to 8 months, and therefore there is a clinical need to reduce the risk of progression after the initial treatment sequence. The REGOBONE study reported a significant PFS benefit of regorafenib (REG) compared to placebo (in osteosarcomas: median PFS: 16.4 versus 4.1 weeks) and a manageable safety profile in patients with histologically confirmed HGBS (i.e., osteosarcoma or other bone sarcomas with the exception of Ewing sarcomas, chondrosarcoma and chordoma). Methods: This multicenter trial is ongoing to study the efficacy and safety of maintenance REG in pts > = 16 years old with HGBS, without complete remission but with no progressive disease after standard treatment, either at diagnosis or at relapse. Sixty pts will be randomly allocated in a 1:1 ratio to receive either oral REG at a daily dose of 120mg or its matching placebo, continuously for a maximum of 12 months. Randomization is stratified according to the setting of the disease: initial diagnosis versus relapse. The primary objective is to compare the efficacy (PFS) between the 2 arms. The expected 4-month PFS rates are 30% in the control arm and 60% in the REGO arm (HR = 0.42). Fifty-two events will provide 87% power to show significant improvement in PFS, using a 2-sided log-rank test at a 5% level. Secondary endpoints include Overall Response Rate (ORR), Disease Control Rate (DCR), Time to Treatment Failure (TTF), Overall Survival (OS), Quality of Life (QoL), and safety profile. Radiological endpoints will be evaluated using the RECIST 1.1 with tumor assessments every 2 months (first 6 months) and then every 3 months. Translational objectives will identify predictive biomarkers for efficacy of REG as maintenance therapy. Pts of the control arm who experience disease progression may switch to receive open label REG. As of Feb 1st, 2022, 3 patients have been randomized. 14 sites of the French Sarcoma Group will participate. An amendment is being implemented to lower the age limit (12 years old) and to expand tumor types to other HGBS (Ewing sarcomas, chondrosarcomas, Undifferentiated Pleomorphic Sarcomas, Leiomyosarcomas and angiosarcomas). Clinical trial information: NCT04698785.

e21601 Background: MBD4 encodes a glycosylase implicated in repair of lesions induced by DNA deamination and its inactivation in tumors is associated with a hypermutated phenotype. Uveal melanoma (UM) is a rare but aggressive form of melanoma carrying an extremely low mutation burden. Although metastatic UM (mUM) are usually resistant to immune checkpoint inhibitors (ICI), the first reported MBD4-mutated (MBD4m) patient responded to ICI, suggesting that MBD4 mutation may predict response to ICI. Since then, other MBD4-inactivated UMs, acute myeloid leukemias, colorectal adenocarcinomas, gliomas and spiradenocarcinoma cases have been reported. Methods: Retrospective cohort of mUM patients treated with ICI. MBD4 was sequenced in a subset of these patients. Results: Three hundred mUM patients were analyzed. Median follow-up was 17.3 months. Ten objective responses and 20 stable disease for > 12 months were observed, corresponding to an objective response rate of 3.3% and a clinical benefit (CB; i.e. responder patients and stable disease) rate of 10%. Median progression-free survivals (PFS) in patients without and with CB were 3.1 and 16.3 months, respectively (p < 0.0001). Of the 134 tumors sequenced for MBD4, five (3.7%) were mutated. MBD4 inactivation was associated with higher mutation burden (179 to 643 variants versus a median of 16 in MBD4 wild-type), better objective response rate as 60% of MBD4m versus 4% of MBD4-wild type patients responded (Fisher’s exact p-test = 0.0009). Median PFS was 4.0 months in MBD4-wild type and 22.3 months in MBD4m patients (HR = 0.22; p = 0.01). Median overall survival was 16.6 months in MBD4-wild type and unreached in MBD4m patients (HR = 0.11; p = 0.004). Conclusions: In mUM patients, MBD4 mutation is highly predictive for response to ICI, PFS and overall survival benefit. MBD4 could be a tissue-agnostic biomarker and should be sequenced in mUM and other tumor types where MBD4 mutations are reported.

11555 Background: Results of the double-blind randomized phase 2 trial (NCT01900743), showed that regorafenib (REG) is an active treatment in patients (pts) previously treated with chemotherapy for an NASTS (cohorts B-leiomyosarcoma, C-synovial sarcoma, D-other sarcoma, Mir 2016), and in pts previously treated with pazopanib (PAZ) (cohort E, Penel 2019). We now present an updated analysis of progression-free survival (PFS) in all NASTS pts to assess the heterogeneity of treatment effect according to histological subtype and prior exposure to PAZ. Methods: Pts received REG 160 mg/d, 21/28 d, or placebo (PB). Pts receiving PB were offered optional cross over in case of centrally confirmed disease progression. The primary endpoint was PFS, according to RECIST-1.1, based on full blinded central review of imaging (including a re-review for cohorts B, C and D, because first analysis was based on a partial central review in these cohorts). Overall survival (OS) was a secondary endpoint. We performed a pooled analysis of cohorts B, C, D and E, on the intent-to-treat dataset, including a multivariate analysis with interaction terms to assess the heterogeneity of treatment effect according to covariates. Results: From 06/2013 to 10/2017, 175 pts were randomized (87 REG vs 88 PB; 56, 27, 55, 37 pts in cohorts B, C, D and E, respectively). The median age was 59 yrs (range, 20-81). There were 101 women (58%). Histological subtype was leiomyosarcoma in 80 pts (LMS; 41 REG vs 39 PB; 56 in cohort B and 24 in cohort E), synovial sarcoma in 28 (SS; 13 REG vs 15 PB; 27 in cohort C and 1 in cohort E), and other sarcoma in 67 (33 REG vs 34 PB; 55 in cohort D and 12 in cohort E). The median number of prior lines of systemic treatment was 2 (range, 1-6). Overall, 43 pts had received prior PAZ (21 REG vs 22 PB). Out of 88 pts assigned to PB, 69 switched to REG after progression (79%). We confirmed a significant PFS-benefit associated with REG in multivariate analysis of the pooled study population, with a HR = 0.48 (95%CI, 0.35–0.66, p < 0.001); median PFS = 2.1 vs 1.0 months, respectively. This benefit appears significant in each histological subtype. However, we observed a borderline interaction between histological subtype and treatment effect (p = 0.09): PFS benefit appears larger in pts with SS (HR = 0.21, 0.10-0.48, p < 0.001) and other sarcoma (HR = 0.49, 0.30-0.81, p = 0.006) than in LMS (HR = 0.59, 0.37-0.93, p = 0.022). PFS benefit appears rather homogeneous across strata of pts with vs without prior exposure to PAZ (interaction test, p = 0.26). Overall, in this study, regorafenib does not show any statistically significant OS-benefit (HR = 0.77, 0.57–1.05, p = 0.10), likely due to the fact that 79% of PB pts crossed over to REG at progression. Conclusions: The present study confirms the clinical PFS benefit associated with regorafenib in all NASTS pts, regardless of prior treatment with pazopanib. Clinical trial information: NCT01900743.

2565 Background: Approval of pembrolizumab for the treatment of unresectable or metastatic MSI-H/dMMR solid tumors that have progressed on prior therapy was supported by data from KEYNOTE-158 (NCT02628067). At the data cutoff of Dec 6, 2018, the ORR was 34.3% among 233 patients (pts) with MSI-H/dMMR solid tumors enrolled in all cohorts of KEYNOTE-158, 77.6% had duration of response (DOR) ≥24 mo, median PFS was 4.1 mo, and median OS was 23.5 mo. We present results from 351 pts enrolled in KEYNOTE-158 cohort K at the data cutoff of Oct 5, 2020. Methods: Cohort K of this phase 2, open-label study enrolled adults with any previously treated advanced noncolorectal MSI-H solid tumor, measurable disease per RECIST v1.1, and ECOG PS of 0–1. MSI-H/dMMR status was assessed locally from a tumor tissue sample and defined as ≥1 of 4 MMR proteins absent by immunohistochemistry or as ≥2 allelic loci size shifts of 5 microsatellite markers by PCR. Pts received pembrolizumab 200 mg Q3W for up to 35 cycles or until PD, unacceptable toxicity, investigator decision, or withdrawal of consent. The primary endpoint was ORR per RECIST v1.1 by blinded independent central review (BICR). Secondary endpoints were DOR and PFS per RECIST v1.1 by BICR, OS, and safety. Efficacy was assessed in all pts who received ≥1 dose of treatment with ≥6 mo follow-up; safety was assessed in all treated pts. Results: 351 pts were enrolled in KEYNOTE-158 cohort K across multiple tumor types, including endometrial (22.5%), gastric (14.5%), small intestine (7.4%), ovarian (7.1%), cholangiocarcinoma (6.3%), and pancreatic (6.3%). 41.0% had 1 prior line of therapy; 55.6% had ≥2 prior lines. Median time from first dose to database cutoff (Oct 5, 2020) was 37.5 (range, 0.2–55.6) mo; 16.0% were continuing treatment. The ORR among the 321 eligble pts was 30.8% (CR, 27; PR, 72); median DOR was 47.5 mo (Table). Treatment-related AEs occurred in 64.7% of pts (grade 3–5, 12.0%), led to discontinuation in 6.6%, and led to death in 3 pts (myocarditis, pneumonia, and Guillain-Barre syndrome). Immune-mediated AEs and infusion reactions occurred in 20.2% of pts (grade 3–4, 4.3%) and led to death in 2 pts with no other contributing factors (myocarditis [AE start, day 26; death, day 33] and Guillain-Barre syndrome [AE start, day 22; death, day 41]). Conclusions: Pembrolizumab demonstrated a high ORR (30.8%), durable clinical benefit, and a manageable safety profile in this heavily pretreated advanced MSI-H/dMMR noncolorectal pan-tumor population. Clinical trial information: NCT02628067. [Table: see text]

e13594 Background: During the French COVID-19 pandemic, clinical activity for cancer care has declined and modified the management of patients in cancer institute. France has imposed two national lockdowns of the population (from March 17 to May 11 and from October 30 to December 15 2020). We evaluate the impact of the lockdown period in North France on clinical activity. Methods: We measured and compared key-indicators of clinical activity between different periods: 2019 versus 2020; before and during the first and second lockdown. Results are given as percentage of changes with 95% confidence interval (CI). Results: The comparative analysis between 2019 and 2020 showed that most clinical activities except hospitalizations (-0.0% [-0.2; +0.1%] has changed. There was a significant increase in overall treated patients (+0.4% [+0.3; +0.5], telemedicine +99.9% [+99.4; +99.7], MRI (+8.8% [+8.2; +9.5], ambulatory chemotherapy (+3.5% [+3.3; +3.8]), number of chemotherapies prepared (+2.0 [+.1.9; +2.2]) and admission in palliative care unit (+15.3%, [+10.8; +19.8]). On the contrary, a significant decrease in treated patients aged > 65 years old (-6.4% [-6.9; -5.9]), incidence rate (-3.1% [-3.6; -2.7]), multidisciplinary team meeting on patient assessment (-3.1% [-3.5;-2.5]), pathology reports (-2.9% (-3.3;-2.6]), radiotherapy (-4.6% [-4.7;-4.4]) and surgery (-6.7% [-7.4;-5.6]) were observed. In 2020, before and during the first lockdown all key-indicators declined except telemedicine (+40,400%) and admission in palliative care unit (+35.5% [+31.0-+49.7]). During the first lockdown, key-indicators decreased from -63.0% [-70.3-59.2] for outpatient visits to -5.1% [-7.0-3.8] for chemotherapy. The impact caused by the second lockdown was non-significant compared to the first one since all key-indicators has increased during the second lockdown: from +23.8% [+22.2; +25.5] for radiotherapy to +76.6% [+72.5; +79.5] for the management of new cases. Conclusions: Dramatic changes occurred in 2020 during the COVID-19-related lockdown compared to 2019. An Increase in chemotherapy treatment and admission on palliative care unit and a decrease in surgery and radiotherapy strongly suggests that the clinical activity was changed in 2020 compared to 2019. This was in line with the reduction of cancer screening in 2020. The impact of the first lockdown on clinical activities was higher compared to the second one. These observations should be investigated further in order to better understand the effects of a pandemic-related lockdown on cancer care.

5046 Background: Darolutamide (Daro) and enzamutamide (Enza) are both next generation androgen receptor inhibitors with demonstrated activity in men with CRPC. Although both agents are associated with survival improvement, their toxicity profiles are different. To help decipher whether this may impact on patient preference, we designed the ODENZA trial. Methods: ODENZA is a prospective, randomized, open-label, multicenter, cross-over, phase II trial of preference between Daro and Enza in men with asymptomatic or mildly symptomatic metastatic CRPC. Patients were randomized 1/1 to receive Daro 1200 mg/d for 12 weeks followed by Enza 160 mg/d for 12 weeks (Daro-Enza arm) or the reverse sequence (Enza-Daro arm). In both arms, the second treatment was given in absence of evidence of cancer progression at week 12. The primary endpoint was patient preference between the two drugs, as assessed by a questionnaire at week 24. The Prescott's test was used to determine treatment preference in patients fullfilling pre planned criteria (exposure to both treatments, no progression at week 12, and completion of the preference questionnaire). A p-value greater than 0.05 indicates that there is no difference in preference between treatments. Stratification factors were performance status and prior taxane for mCSPC. After week 24, patients went on to an extension period during which they received the chosen treatment until progression or toxicity. The main secondary objectives included reasons for preference, response at week 12, cognitive assessment, and toxicity. Results: Overall 249 pts were randomized, median age 72y (68; 79), ECOG PS 0 (56%), prior taxanes (22%). Two hundred pts fulfilled the pre-planned criteria for evaluation of the preference primary endpoint : 97 (48.5% [41.3;55.7]), 80 (40.0% [33.0;47.0]), and 23 (11.5% [6.8;16.2]) chose Daro, Enza, and had no preference, respectively (unilateral p-value of 0.92). After preference assessment, 186 patients entered the extension period: 103 (55.4%) and 83 (44.6%) received Daro and Enza respectively. The most common factors influencing patient preference all numerically favored Daro over Enza, without significant differences were: less fatigue (44% vs 29%), ease of taking the medication (37% vs 31%), better quality of life (36% vs 28%), ability to be more active (26% vs 15%), ability to concentrate (22% vs 15%) and less falls (6% vs 3%). A PSA50 response was achieved in 76.2% and 83.9% at week 12 with Daro and Enza respectively (p = 0.13). Fatigue was the most frequently reported all grade adverse event at week 12, in 21% and 36% with Daro and Enza, respectively. Conclusions: More patients with early mCRPC preferred Daro over Enza, although the difference did not reach significance, with fatigue as the key influencing factor. Clinical trial information: NCT03314324.

e24074 Background: Admission of cancer pts in ICU is a medical and ethical challenge, since there is no reliable prognostic tool for guiding decision making. Therefore, recent progress in cancer management improved the OS of advanced cancer pts, rendering this issue more and more frequent. Methods: We have retrospectively analyzed the medical charts of 129 consecutive pts treated in our institution and admitted in ICU from 01/2014 and 04/2019. We identified PF using Cox Models. We analyzed PF for OS or for ability to restart systemic treatment. Results: At the time of ICU admission, the mean age was 58.9 (range, 25-81). The were 51% of men. PS were 0-1 in 61% and 2-3 in 39%. The most prevalent cancers were lung (20%), sarcoma (17%), breast (16%) and gynecological cancers (11%). The number of metastatic sites was 1 (17%), 2 (39%), 3 (26%) and more (19%). The malignancy was stable in 69% and progressive in 31%. Pts currently received systemic treatment in 78%, were free of treatment in 5% and were not yet treated in 16%. Cancer itself (53%), toxicity of treatment (43%) and underlying comorbidities (37%) contributed to ICU admission. The number of organ failures was 0 in 12%, 1 in 40%, 2 in 26% and 3 or more in 22%. The 4 most symptoms were dyspnea (34%), severe infection (25%), cardiac event (23%) and bleeding (10%). Pt management required vasoamines administration (38%), ventilation (31%) and dialysis (4%). The median duration of stay in ICU was 4 days (range, 0-71). 21 (16%) pts died in ICU and 16 (12%) just after ICU discharge. 83 pts returned to home (64%). Systemic treatment was done in 61% of pts who left hospitalization alive after ICU. 1-month, 6-month and 12-month OS were 67, 36 and 21%, respectively. Multivariate analysis identified the following PF for OS: PS = 0 (HR = 0.57 [0.38-0.79]; p = .007), admission related to underlying comorbidity (HR = 0.63 [0.41-0.98], p = .0001) or admission related to treatment toxicity (HR = 0.31 [0.17-0.56] p = .0001). Multivariate identified 1 PF for ability to restart systemic treatment: one or less organ failure (HR = 0.34 [0.14-0.83], p = 0.02). Conclusions: In ICU, OS of adult pts with solid tumors looks like the non-cancer population. ¼ of admitted pts died in ICU or just after. Key PF were PS and cause of critical state. When ICU admission is related to cancer itself, the prognosis was poor.

3096 Background: DPD deficiency screening before fluoropyrimidine is a matter of debate. To avoid lethal toxicity, French authorities impose DPD screening before fluoropyrimidine-based chemotherapy by dosing uracilemia since April 2019. Methods: We have included all consecutive adult patients receiving 5-fluorouracil (5-FU) or capecitabin from April 2019 to January 2020 in 6 cancer centers. During the study period, different methods for screening had been applied: DPYD complete sequencing, phenotype (uracilemia and/or dihydrouracilemia/uracilemia ratio - UH2/U -) or both. All sceening tests were conducted in the same laboratory. Association between the method of DPD screening and fluoropyrimidine dose reduction at second chemotherapy cycle or on severe ≥grade 3 early toxicity (between first and second cycle) was evaluated using multivariate logistic regression. Concordance between genotype and phenotype for DPD deficiency was explored using Cohen Kappa test. Results: We included 597 patients, the median age was 63 (range, 55-77). The most prevalent cancers were digestive (68.3%), head and neck (19.4%) and breast (9.2%). 12.3% of patients received capecitabine and 87.3% received polychemotherapy. DPD deficiency screening was done for most of patients (n=519, 86.9%). DPD screening method consisted in full sequencing of DPYP (n=41; 7.9%), phenotype analysis (n=44, 8.5%) or both (n=424, 83.6%). We did not identify any complete DPD deficiency. Uracilemia was dosed for 467 patients, the median was 6.5 ng/mL and for 21 patients (4.5%) uracilemia was > 16 ng/mL and/or UH2/U

6020 Background: Bintrafusp alfa is a first-in-class bifunctional fusion protein composed of the extracellular domain of the TGF-βRII receptor (a TGF-β “trap”) fused to a human IgG1 mAb blocking PD-L1. A previous report of an expansion cohort from a phase 1 study (NCT02517398) suggested that bintrafusp alfa had a manageable safety profile and early signs of clinical activity in patients with heavily pretreated, advanced SCCHN after a median follow-up of 86.4 weeks. Here we report long-term efficacy and safety for this cohort. Methods: Patients with advanced SCCHN that progressed/recurred after platinum therapy in the recurrent/metastatic setting, or < 6 months after platinum therapy in the locally advanced setting, received bintrafusp alfa 1200 mg every 2 weeks until confirmed progressive disease, unacceptable toxicity, or trial withdrawal. The primary endpoint was confirmed best overall response assessed per RECIST 1.1 assessed by independent review committee (IRC); safety was a secondary endpoint. Results: As of May 15, 2020, 32 patients had received bintrafusp alfa for a median of 2.8 months (range, 0.5-29.9 months), no patient remained on treatment, and median follow-up to data cutoff was 41.7 months (range, 39.8-43.5 months). The objective response rate (ORR; 13%) was unchanged since the previous report; median duration of response (DOR) was increased at 21.4 months (95% CI, 5.5 months to not reached [NR]). While the clinical activity of bintrafusp alfa may be improved in patients with HPV-positive tumors (Table), outcomes were generally similar between PD-L1 subgroups (≥1% vs < 1% tumor cells). The overall safety profile was consistent with the previous report for this cohort, without grade 4 nor 5 treatment-related adverse events (TRAEs); no new TRAEs of grade 3 or that led to discontinuation of bintrafusp alfa were reported. Conclusions: With a median follow-up of over 3 years in patients with heavily pretreated advanced SCCHN, bintrafusp alfa showed sustained clinical activity and 3-year OS of 24.0%, which compares favorably to historical data. Clinical activity appeared to be greater in patients with HPV-positive tumors than those with HPV-negative tumors. The safety profile was manageable and consistent with earlier analysis. Further investigation of bintrafusp alfa in SCCHN and other HPV-associated cancers is ongoing. Clinical trial information: NCT02517398. [Table: see text]

11520 Background: AcSé Pembrolizumab is a Phase 2, non-randomized parallel arms, open-label, multicentric study from Unicancer investigating the efficacy and safety of pembrolizumab monotherapy in different cohorts of patients with rare cancers (NCT03012620). Here we report the results of pembrolizumab in the rare sarcoma cohort. Methods: Selected histotypes were all rare sarcomas patients (pts) (incidence < 0.2/100,000/year). Main inclusion criteria were age > 18, ECOG PS≤1 and advanced or metastatic disease resistant to standard treatment. Patients received pembrolizumab 200 mg IV as a 30-minute infusion on Day 1 of every 21-day cycle for a maximum of 2 years. The primary endpoint was the confirmed objective response rate according to RECIST v1.1 at 12 weeks. Secondary endpoints included clinical benefit rate, duration of response, progression-free survival (PFS), overall survival (OS), and safety. Five groups of pts were distinguished, namely chordoma, alveolar soft-part sarcoma (ASPS), desmoplastic small round cell tumor (DSRCT), smarca4 deficient malignant rhabdoid tumor (SMRT), and other histotypes. Results: 98 patients including 34 with chordoma, 14 ASPS, 11 SMRT, 8 DSCRT and 31 with other histotypes, were included from July 2017 to December 2020. The median number of cycles was 5 (range, 1 to 35) with 78 (79.6%) patients who discontinued the trial after a median of 4 cycles. There were 6 (7.3%) partial response (PR) at 12 weeks. The best response was CR in 1 patient (1%), PR in 14 patients (14.3%), and stable disease (SD) in 33 (33.7%). Median duration of response was 8.2 months [IQR, 4.1 to 9.0]. The occurrence of best response depended on the histotype, with 3 (8.8%) responses in chordoma, 7 (50%) in ASPS, 3 (27%) in SMRT, 1 (12.5%) in DSCRT and 1 (3.2%) in other histotypes (p = 0.0011). At the data cut off, median PFS was 2.75 months, and median OS was 19.7 months on the overall population. Outcomes differed according to the histotype group, with the 12 months PFS rates at 31.2% (chordoma), 35.7% (ASPS), 18.2% (SMRT), 0% (DSCRT) and 3.3% (other), respectively (p < 0.0001), and median PFS at 6.6 (chordoma), 7.5 (ASPS), 1.1 (SMRT), 2.1 (DSCRT) and 2.1 months (other), while 1-year OS rates were 76.6% (chordoma), 85.7% (ASPS), 36.4% (SMRT), 17.5% (DSCRT) and 42.9% (other) with median OS only reached for SMRT (2.4 months), DSRCT (10 months), and the other histotype group (7.1 months) (p = 0.004). The side effect profile of pembrolizumab was similar to other tumor type. Conclusions: Pembrolizumab is safe and well tolerate in this pop od STS pts, AcSé study reports high levels response rate and prolonged activity in selected subtypes of rare sarcomas. Clinical trial information: NCT03012620.

TPS11576 Background: Pts with OS and non-OS of bone are treated with a multimodal sequence therapy of neoadjuvant chemotherapy (CT), surgery and adjuvant CT, followed by a close surveillance until recurrence. At recurrence, the prognosis remains poor with objective response rates of 3-29%, and a median Progression-Free Survival (PFS) of less than 4 months in OS. There is a clinical need to reduce the risk of recurrence after the initial treatment sequence. The REGOBONE study reported a significant clinical benefit of regorafenib compared to placebo in patients with relapsed OS (median PFS: 16.4 versus 4.1 weeks). Methods: This multicenter trial is ongoing to study the efficacy and safety of maintenance REGO in pts > = 16 years, with complete remission after initial treatment sequence of their bone sarcoma. 168 pts will be randomly allocated in a 1:1 ratio to receive either oral REGO or its matching placebo (control arm) at a daily dose of 120mg, continuously and for a maximum of 12 months. Randomization will be stratified according to the following risk factors: metastases (mets) at diagnosis and/or poor response to neoadjuvant CT versus no mets at diagnosis and good response to neoadjuvant CT. The primary objective is to compare the efficacy (Relapse-Free Survival) between the 2 arms. The expected 3-year RFS rates are 55% in the control arm and 74.6% in the REGO arm (HR = 0.5). 66 events will provide 80% power to show significant improvement in RFS, using a 2-sided log-rank test at a 5% level. Secondary endpoints include Time to Treatment Failure, Overall Survival, Quality of Life, safety profile, and compliance to treatment. Radiological endpoints will be evaluated using the RECIST 1.1. Translational objectives will be to identify predictive biomarkers for efficacy of REGO as maintenance therapy using liquid biopsies. As of Feb 1st, 2021, 3 patients have been randomized. 15 sites of the French Sarcoma Group will participate. Clinical trial information: NCT04055220.

11510 Background: DF rare locally aggressive fibroblastic non-metastasizing tumor, with an unpredictable course. Its management remains challenging, there is a current shift in standard of care from large surgical resection (SR) to active surveillance (AS). Most of DF display somatic mutation of CTNNB1, with three major hotpots: S45F, T41A and S45P. The poor prognosis of S45F is a matter of debate (Timbergen et al. Ann Surg 2019). Methods: ALTITUDES (NCT02867033) is a nationwide prospective registry of DF, diagnosed from January 2016 to December 2020 and confirmed by central pathological review. CTNNB1 mutations were identified by NGS. Primary endpoint was event-free survival (including disease progression or relapse). We have selected pts managed by AS, SR or systemic treatments as front-line. Pt undergoing R2 resection and then managed by follow-up were part of AS group. Prognostic factors were assessed using univariate and multivariate Cox Model. Results: From the 630-pts enrolled in ALTITUDES, 500 (79.3%) were eligible for the present analysis. Exclusion criteria were diagnosis before 2016 in 13 pts, multiple DF in 33 pts, 39 pts without CTNNB1 mutation analysis, and 45 pts receiving other treatments. The study population included 349 females (69.8%), the median age was 40 years (range 1-89). Abdominal wall was the predominant primary site: 161 pts (32.2%). In 430 (86.0%) cases, there was a CTNNB1 mutation, including, S45F in 56 cases (11.2%). In 70 cases (14.0%), we did not identify CTNNB1 mutation. The front-line managements were AS in 361 pts (72.2%), SR with R0/R1 margins in 57 cases (11.4%) and systemic treatments in 82 pts (16.5%). The median follow-up was 23 months (Range, 0.4-55). Overall, progression or relapse occurred in 128 pts (25.6%). We observed a significant EFS-difference between treatment groups, both in univariate and multivariate analysis with, compared to AS, a better outcome in patients with SR and worse outcome in patients who had received a systemic treatment (p = 0.01 in multivariate analysis). The risk of event was significantly associated with the tumor size, with a HR = 1.46 in tumors larger than 50 mm compared to smaller tumors (95%CI, 1.01-2.10, p = 0.04). We did not observe any significant association between the CTNNB1 mutational status and the outcome: compared to patients with another mutation, the hazard ratio associated with a S45F mutation was HR = 0.84 (95%CI, 0.48-1.46, p = 0.53) in multivariate analysis. Age, gender, location (abdominal wall versus other) were not associated with EFS. Conclusions: In this large prospective study, S45F was not an independent poor prognostic factor in DF. Size and front-line treatment drive both the outcome. The understanding and prediction of natural course of DF require further studies.

e13604 Background: To limit SARS-Co-2 transmission in general population, French government had set up a first National containment from March 16, 2020 to June 2, 2020. Furthermore, General Direction of our hospital have implemented and organized physical distancing (telework, teleconsultation, virtual multi-disciplinary board), mask wearing, use of alcoholic ... to limit interpersonal contacts. To assess the impact of this policy, we have carried out a seroprevalence study and identified risk factors for SARS-Co-2 prevalence among HCP in May/June 2020. Methods: This is part of CORSER Study (“Etude séro-épidémiologique du virus SARS-CoV-2 en France d’une Cohorte CORSER-2d de personnels d’établissements de santé »), registered with ClinicalTrials.gov (NCT04325646). Two serological tests were applied in this cohort: S-Flow assay and Luciferase-linked immune-sorbent assay (LuLisa-N test). Results: Between 04 May to 26 June 2020, 392 Osar Lambret Cancer Center employees were included in the present study (about 40% of all staff). The most common jobs were: 92 nurses (23.4%), 80 radiology/radiotherapy manipulators (20.4%), 51 physicians (13.0%), 33 medical interns (8.4%) and 47 administrative staff members (11.9%). There were 98 men (25.0%) and 294 women (75.0%). The mean age was 38.7 (+/- 11.4). There were 7 seropositive cases; the seroprevalence was 1.8% (95%-CI: 0.7-3.6). Among the 7 positive cases, 5 were symptomatic (71.4%). In univariate analysis, factors associated with SARS-Co-2 seroprevalence: symptoms suggesting viral infection within 2 months (OR=5.33), dysgueusia (OR=37.00), anosmia (OR=66.29) and HCP exposed to COVID-19 patient outside work (OR=6.69). Gender, tobacco consumption, O blood group, HCP versus administrative staff, HCP working in different services, HCP providing care to suspected COVID-19 patients, HCP providing high-risk cares were not found to be associated with seropositivity. Of note, we have noticed that BMI≥ 24 kg/m² was associated with seroprevalence (OR=15.45), without biological rational. Conclusions: Our study suggests that strict implementation of protective measures was associated with low SARS-Cov-2 prevalence at the end of first National Containment, including among HCP treating COVID-19+ patients. HCPs seroprevalence seemed lower than seroprevalence of the general population at the same period (4-5%). Clinical trial information: NCT04325646.

e24026 Background: Despite largely diffused guidelines, Pneumococcal and Influenza vaccination coverage (VC) remains insufficient in cancer patients receiving chemotherapy. We evaluated VC and carried out an interventional study to improve VC in cancer patients treated in 3 medical oncology departments in 3 hospitals of North-of-France. Methods: Using a standardized questionnaire, we assessed the VC in adult cancer patients receiving anticancer treatment in day hospital between 02 and 07 Dec. 2019. Then, in each hospital, we organized trainings with physicians to discuss the current vaccination guidelines (January 2020). Finally, to assess the impact of this intervention on Pneumococcal and Influenza VC, we have conducted a similar survey using the same questionnaire in March 2020. Since there were no specific guidelines on Diphtheria-Tetanus-Pertussis (DTP) VC and we did not expect an improvement, we have monitored DTP VC as internal control. Results: 272 patients have been enrolled in the “before study” in the 3 hospitals, whereas 156 patients have been enrolled in the “after study” in only 2 hospitals (the after-study was not feasible in 3rd hospital because of COVID-19 pandemic national containment). Predictors associated with DTP VC was age, with Pneumococcal VC was center and with Influenza VC were age, gender and tumor histology (adenocarcinoma vs others). Influenza VC was significantly improved after intervention (42.6 vs 55.1% with p = 0.016) especially in fragile patients, but Pneumococcal VC was not (11.8 vs 15.4% with p = 0.357). Conclusions: As expected, VC was very low in cancer patients, our figures are consistent with literature data. The impact of intervention (training of physicians) is limited without improvement of Pneumococcal VC. We presume that this increase in Influenza VC mainly reflects the overall result of national Influenza vaccination campaign.

4080 Background: Second-line treatment options for patients with biliary tract cancers (BTC) are limited. Lenvatinib, an anti-angiogenic multikinase inhibitor, in combination with the programmed death-1 immune checkpoint inhibitor pembrolizumab, has demonstrated promising antitumor activity with a manageable safety profile in patients with select advanced solid tumors. LEAP-005 (NCT03797326) is evaluating the efficacy and safety of lenvatinib plus pembrolizumab in patients with previously treated advanced solid tumors; here we present results from the BTC cohort of LEAP-005. Methods: In this nonrandomized, open-label, phase 2 study, eligible patients were aged ≥18 years with histologically or cytologically documented advanced (metastatic and/or unresectable) BTC with disease progression after 1 prior line of therapy, measurable disease per RECIST v1.1, ECOG PS of 0‒1, and tissue sample evaluable for PD-L1 expression. Patients received lenvatinib 20 mg once daily plus pembrolizumab 200 mg Q3W for up to 35 cycles (approximately 2 years) or until confirmed disease progression, unacceptable toxicity, or withdrawal of consent. Treatment with lenvatinib could continue beyond 2 years in patients experiencing clinical benefit. Primary endpoints were ORR (per RECIST v1.1 by blinded independent central review) and safety. Secondary endpoints were the disease control rate (DCR; comprising CR, PR, and SD), duration of response (DOR), PFS, and OS. Tumor imaging was performed Q9W from treatment initiation for 54 weeks, then Q12W to week 102, and Q24W thereafter. Results: 31 patients were enrolled in the BTC cohort (ECOG PS 1, 55%; 84% ex-US). As of April 10, 2020, median time from first dose to data cutoff (DCO) was 9.5 months (range, 3.1‒11.9), with 8 patients on treatment at DCO. There were 3 (10%) PRs and 18 (58%) SDs. ORR was 10% (95% CI, 2‒26), and DCR was 68% (95% CI, 49‒83). Median DOR was 5.3 months (range, 2.1+ to 6.2). Median PFS was 6.1 months (95% CI, 2.1‒6.4). Median OS was 8.6 months (95% CI, 5.6 to NR). Treatment-related AEs occurred in 30 patients (97%), including 15 (48%) who had grade 3 AEs; there were no grade 4 or 5 treatment-related AEs. 2 (6%) discontinued treatment due to treatment-related AEs (myocarditis, pyrexia; n = 1 each). The most frequent treatment-related AEs were hypertension (42%), dysphonia (39%), diarrhea (32%), fatigue (32%), and nausea (32%). 14 patients (45%) had immune-mediated AEs and 1 patient (3%) had an infusion-related reaction. Conclusions: Lenvatinib plus pembrolizumab demonstrated encouraging efficacy and manageable toxicity in patients with advanced BTC who had received 1 line of prior therapy. Based on these data, enrollment in the BTC cohort has been expanded to 100 patients. Clinical trial information: NCT03797326.

11507 Background: PD1 inhibition has shown limited activity in all comers clinical trials including patients with advanced soft-tissue sarcomas (STS). In the PEMBROSARC study, objective response rate, progression-free (PFS) and overall survival (OS) were respectively 2.1%, 1.4 and 7.1 months respectively (Toulmonde et al. Jama Oncol 2017). We have recently shown that the presence of tertiary lymphoid structures (TLS) may represent a biomarker to select patients who are more likely to benefit from immunotherapy (PetitPrez et al., Nature 2020). We report here the first clinical trial investigating the efficacy of PD1 inhibition in TLS-positive STS. Methods: PEMBROSARC is an open-label multicenter phase II study of pembrolizumab in combination with low-dose cyclophosphamide in pts with STS selected based on the presence of TLS. TLS status has been assessed centrally has previously described (PetitPrez et al., Nature 2020). Eligible patients received pembrolizumab 200mg IV q21 days and cyclophosphamide 50 mg BID 1week on, 1 week off. All patients had confirmed progressive disease at inclusion based on central review of two imaging performed at less than 6 months interval. The primary efficacy endpoint was 6-month non-progression (as per RECIST evaluation criteria v1.1). Based on the following hypotheses: 15% 6-month non-progression rate (H0), 40% acceptable 6-month non-progression rate (H1), 5% type I error rate, 90% power, a total of 29 assessable patients were necessary and 8 patients or more had to be progression-free at 6 months to reach the first endpoint. Results: 240 patients were screened for TLS status between September 2018 and January 2020 in 7 centers of the French Sarcoma Group. Among them, 48 were found to be TLS+ as per central review and 35 were included in the study. The three most frequent histological subtypes were: well-differentiated/dedifferentiated liposarcoma (n = 13); UPS (n = 6), and leiomyosarcoma (n = 4). 30 patients were eligible for efficacy. Of those, as per central imaging review, 13 patients (43.3%) had tumor shrinkage resulting in partial response in 8 patients (26.7%) and stable disease in 5 cases (16.7%). 10 patients had progressive disease. Twelve patients were progression-free at 6 months (40.0% 95%CI = [22.7 – 59.4]).Median PFS and OS were 4.1 months (95%CI, 1.4-9.6) and 14.5 months (95%CI, 8.5- 18.3 months), respectively. Conclusions: With an objective response and a 6-month non-progression rates of 26.7% and 40% respectively versus 2.1% and 4.2% in all comers, the PEMBROSARC study confirms that selection based on TLS status is an efficient approach to tailor immunotherapy in STS patients. Clinical trial information: NCT02406781.

126 Background: The Phase 3 PROfound trial (NCT02987543) met its primary endpoint and key secondary endpoints, including improved overall survival (OS) for olaparib in men with mCRPC with alterations in BRCA1, BRCA2, or ATM (Cohort A). We report gene-by-gene analysis of olaparib antitumor activity among the 15 prespecified homologous recombination repair (HRR) genes. Methods: Pts were randomized to olaparib (300 mg bid; n=256) or physician’s choice of enzalutamide or abiraterone (control; n=131). Exploratory analyses in pts with alterations in BRCA1 and/or BRCA2 (BRCA, regardless of co-occurring alterations with other HRR genes) or in single genes were conducted. Results: Evidence of olaparib antitumor activity was observed in subgroups with >10 pts (table). Data for pts with alterations in only BRCA1, BRCA2, PPP2R2A, RAD51B, RAD54L, PALB2, BRIP1, CHEK1, BARD1, and RAD51D will be reported (no FANCL or RAD51C enrolled). Conclusions: Small subgroups limit interpretation for some genes. Olaparib antitumor activity is greatest in pts with BRCA alterations, with a spectrum of clinical sensitivity to olaparib as defined by rPFS and OS across the broader population with alterations in other HRR genes. Clinical trial information: NCT02987543. [Table: see text]

321 Background: Second-line treatment options for patients with biliary tract cancers (BTC) are limited. Lenvatinib, an anti-angiogenic multikinase inhibitor, in combination with the programmed death-1 immune checkpoint inhibitor pembrolizumab, has demonstrated promising antitumor activity with a manageable safety profile in patients with select advanced solid tumors. LEAP-005 (NCT03797326) is evaluating the efficacy and safety of lenvatinib plus pembrolizumab in patients with previously treated advanced solid tumors; here we present results from the BTC cohort of LEAP-005. Methods: In this nonrandomized, open-label, phase II study, eligible patients were aged ≥18 years with histologically or cytologically documented advanced (metastatic and/or unresectable) BTC with disease progression after 1 prior line of therapy, measurable disease per RECIST v1.1, ECOG PS of 0‒1, and tissue sample evaluable for PD-L1 expression. Patients received lenvatinib 20 mg once daily plus pembrolizumab 200 mg Q3W for up to 35 cycles (approximately 2 years) or until confirmed disease progression, unacceptable toxicity, or withdrawal of consent. Treatment with lenvatinib could continue beyond 2 years in patients experiencing clinical benefit. Primary endpoints were ORR (per RECIST v1.1 by blinded independent central review) and safety. Secondary endpoints were the disease control rate (DCR; comprising CR, PR, and SD), duration of response (DOR), PFS, and OS. Tumor imaging was performed Q9W from treatment initiation for 54 weeks, then Q12W to week 102, and Q24W thereafter. Results: 31 patients were enrolled in the BTC cohort (ECOG PS 1, 55%; 84% ex-US). As of April 10, 2020, median time from first dose to data cutoff (DCO) was 9.5 months (range, 3.1‒11.9), with 16 patients on treatment at DCO. There were 3 (10%) PRs and 18 (58%) SDs. ORR was 10% (95% CI, 2‒26), and DCR was 68% (95% CI, 49‒83). Median DOR was 5.3 months (range, 2.1+ to 6.2). Median PFS was 6.1 months (95% CI, 2.1‒6.4). Median OS was 8.6 months (95% CI, 5.6 to NR). Treatment-related AEs occurred in 30 patients (97%), including 15 (48%) who had grade 3‒4 AEs; there were no treatment-related deaths. 2 (6%) discontinued treatment due to treatment-related AEs (myocarditis, pyrexia; n = 1 each). The most frequent treatment-related AEs were hypertension (42%), dysphonia (39%), diarrhea (32%), fatigue (32%), and nausea (32%). 14 patients (45%) had immune-mediated AEs and 1 patient (3%) had an infusion-related reaction. Conclusions: Lenvatinib plus pembrolizumab demonstrated encouraging efficacy and manageable toxicity in patients with advanced BTC who had received 1 line of prior therapy. Based on these data, enrollment in the BTC cohort has been expanded to 100 patients. Clinical trial information: NCT03797326.

3089 Background: In Study JVDF (NCT02443324), we combined ramucirumab (VEGFR2 antagonist) and pembrolizumab (PD-1 antagonist) to simultaneously target the tumor microenvironment and immune checkpoints in patients with advanced non-small cell lung cancer (NSCLC), gastric or gastroesophageal junction adenocarcinoma (G/GEJ), urothelial carcinoma (UC) or biliary tract cancer (BTC). We reported that this combination was associated with increased antitumor activity in patients with PD-L1 positive tumors by immunohistochemistry (IHC) compared with PD-L1 negative tumors.1) Here we explore the association between baseline gene expression profiles and clinical outcomes. Methods: JVDF was a nonrandomized phase 1a/b trial that treated patients with intravenous ramucirumab at 8 mg/kg on days 1 and 8 (G/GEJ, BTC) or 10 mg/kg on day 1 (G/GEJ, NSCLC, UC) plus pembrolizumab (200 mg day 1) every 3 weeks.1 Baseline tumor samples from 53 patients across 7 study cohorts were analyzed with the NanoString PanCancer Immune Profiling Panel for RNA expression and the DAKO PD-L1 IHC 22C3 pharmDx assay for PD-L1 protein expression. Clinical outcomes included progression-free survival (PFS), overall survival (OS), and objective response rate (ORR). Results: Across cohorts, PD-L1 gene expression was correlated with increased IFNγ gene expression and immune-related gene signatures (T effector, T cell-inflamed (TIS)), and trended with PD-L1 protein expression. Expression of immune checkpoint-related genes and myeloid-derived suppressor cell /regulatory T cell markers was increased in the NSCLC TPS≥50% PD-L1 IHC subgroup (N=8), while no clear pattern of expression was observed in other cohorts. Higher T effector and TIS scores appeared associated with better survival and response in NSCLC cohorts (mean TIS: 1.21±0.80 in responders (N=7) vs -0.13±0.57 in non-responders (N=7); p=0.004), and a trend was observed in G/GEJ cohorts (mean TIS: -0.18±0.30 (N=5) vs -0.39±0.21 (N=13); p=0.207). Of note, partial responses and stable disease were also observed in NSCLC and G/GEJ patients with a low baseline inflammatory signature score. Additional analyses are ongoing and will be presented. Conclusions: Baseline PD-L1 gene and protein expression tends to correlate with immune-related gene expression, and an inflamed tumor microenvironment may be associated with better clinical outcomes with ramucirumab and pembrolizumab. However, interpretation is limited by lack of control arm and sample size.1) Herbst et al. Lancet Oncol. 2019 Aug; 20 (8):1109-1123. Clinical trial information: NCT02443324 .

e23558 Background: DF are locally invasive soft tissue tumors of fibroblast origin, with unpredictable course. CTTNB1 mutations are a hallmark of DF. Molecular-targeted therapies are lacking. Recently developed OncoSignal (www.philips.com/oncosignal) pathway activity tests enable quantitative measurement of activity of clinically relevant oncogenic signal transduction pathways (estrogen, progesterone & androgen receptors; PI3K, MAPK-AP1, TGFβ, Notch, Hedgehog, Wnt pathways), based on computational interpretation of expression levels of direct target genes of the pathway-associated transcription factor. Methods: OncoSignal was used to measure activity of signaling pathways on Affymetrix transcriptome data from sporadic DF (n = 128), generated before in our center (GEO database: GSE58697), and from presumably healthy fibroblasts (n = 63) derived from a number of different tissue types (public dataset GSE63626). Pathway activity was expressed on a normalized scale 0-100 (min and max pathway activity score depend on the cell/tissue type analyzed). Comparisons used Mann Whitney test. Results: Mean Notch (43.2+/-7.2 vs 11.2 +/- 5.2; p < .0001), PI3K (54.6 +/- 10.0 vs 48.1 +/- 6.6, p < .0001) and androgen receptor (AR) pathway activities (32.8 +/- 3.7 vs 21.7 +/- 2.4, p < .0001) were increased in DF compared to fibroblasts. MAPK-AP1 pathway activity was lower in DF (40.8+/-6.0 vs 70.2 +/- 9.5; p < .0001). Mean Wnt (p = 0.7), HH (p = 0.8) pathway activities were not different between DF and fibroblasts, however the range in Wnt pathway activity scores was larger in DF (median score 24.1, range 5.3-41.7) than in normal fibroblasts (median 26.0, range 14.6-33.0). Conclusions: The increased activity of Notch, PI3K pathways in DF is consistent with the mesenchymal nature. We confirm here the key role of AR pathway. The wide range of Wnt pathway expression was larger, suggesting the heterogeneity in TD biology, possibly related to the type of CTTNB1 mutations. Overall, the large inter-patient variation in pathway activities is in line with variety of TD course. We identified signaling pathways which may provide novel options for targeted drug treatment.

11560 Background: Since 2010, presentation to a designated sarcoma tumor board and pathological review by an expert network are mandatory for sarcoma patients in France. NETSARC+ (merging the 3 initial RREPS, RESOS & NETSARC) collected prospectively all cases of reviewed sarcomas and tumors of intermediate malignancy (TIM) nationwide. We report on the incidence of subtypes according to WHO classification from 2013 to 2016. Methods: Sarcoma expert pathologists reviewed samples were all prospectively integrated in the database; the results using the latest WHO classification are presented for the years 2013 to 2016, including yearly variations. Correlation of the incidence of each histotype with dedicated published clinical trials was conducted. Results: 139 different histological subtypes are reported among the 25172 patients with sarcomas (n = 18710, 64%) or TIM (n = 6460, 36%), respectively n = 5838, n = 6153, n = 6654, and n = 6527 yearly from 2013 to 2016. Over these 4 years, the observed yearly incidence of sarcomas, TIM, and all was therefore 79.7, 24.9 and 95.1/10e6/year, above that previously reported. GIST, liposarcoma, leiomyosarcomas, undifferentiated sarcomas represented 13%, 13%, 11% and 11% of all sarcomas. Only GIST, as a single entity exceeded a yearly incidence above 10/million per year. There were respectively 30, 63 and 66 different histological subtypes of sarcomas or TIM (single entities or lumped together, e.g. MPNST, or vascular sarcomas...) with an incidence ranging from 10 to 1/10e6/year, 1-0.1/10e6 per year, or < 0.1/10e6/year respectively. The 2 later “incidence groups” included 21% of the patients. The incidence of 8 histotypes varied significantly over this 4 years. Patients with tumors with an incidence above 1/10e6 per year have significantly higher numbers of dedicated published phase III and phase II clinical trials (p < 10e-6). Conclusions: This nationwide registry of sarcoma patients with an histology reviewed by sarcoma experts shows that the incidence of sarcoma and TIM is higher than previously reported, may vary over years for some histotypes, and that tumors with an incidence < 10e6 have a much lower access to clinical trials.

11506 Background: U-LMS and ST-LMS are rare tumors with poor prognosis when locally advanced or metastatic, and with moderate chemosensitivity. Overall response rates (ORR) given in the 1st-line setting do not exceed 50% for U-LMS and 35% for ST-LMS with a mean response duration of 3- 6 months without impact on overall survival (OS). In 2015 we reported very encouraging results of the LMS-02 study (NCT02131480) with ORR of 59.6% in U-LMS, and 39.3% in ST-LMS with manageable toxicity (Pautier; Lancet oncol 2015). Herein, we report the updated results of progression-free survival (PFS) and final results of overall survival (OS). Methods: Patients (pts) received 60 mg/m² intravenous Doxo followed by trabectedin 1.1 mg/m2 as a 3-hour infusion on Day 1 and pegfilgrastim on Day 2, repeated every 3 weeks for up to 6 cycles. Surgery for residual disease was permitted. Patients were stratified into U-LMS and ST-LMS groups. Results: Overall, 108 patients with LMS with a median age of 59 years and mostly metastatic disease (85%) were enrolled. Of those, 77 patients (71.3%) have received all 6 cycles of treatment, and 20 patients (18.5%) had metastasis resection. With a median follow-up of 7.2 years (95% CI: 6.9 - 8.2), the overall median PFS was 10.1 months (95% CI: 8.5 - 12.6), being 8.3 months (95 CI: 7.4 - 10.3) and 12.9 months (95% CI: 9.2 - 14.1) in U and ST group, respectively. Median OS was 34.4 months (95% CI: 26.9 - 42.7), being 27.5 months (95% CI: 17.9 - 38.2) in U-LMS and 38.7 months (95% CI: 31.0 - 52.9) in ST-LMS group. The median OS among the 20 pts with surgery was not reached vs 31.6 months in the population without surgery (95% IC: 23.9 - 35.4). Conclusions: The Doxo +Trab combination is an effective 1st-line therapy for pts with LMS, with promising PFS and OS results and an acceptable safety profile. Merely for comparison, the most recent results of Doxo alone in metastatic LMS, given in 1st-line setting in a phase III ANNOUNCE trial conducted during the same period, reported median PFS of 6.9 months, and median OS of 21.9 months (ASCO 2019 LBA3). Results of the LMS04 trial (NCT02997358), a randomized phase III study comparing this combination vs Doxo alone in 1st-line therapy in metastatic LMS are pending. Clinical trial information: NCT02131480 .

527 Background: Aberrant activation of the fibroblast growth factor receptor (FGFR) pathway is implicated in many cancers, including UC. In a recent Phase I dose-escalation study, rogaratinib, an oral pan-FGFR1-4 inhibitor, demonstrated favorable efficacy and safety in pts with solid cancers selected based on FGFR1-3 mRNA overexpression. We report results from the Phase I expansion cohort with rogaratinib in pts with UC selected by FGFR1-3 mRNA overexpression and/or FGFR3-activating mutations (NCT01976741). Methods: Pts with advanced/metastatic UC were screened for FGFR1-3 mRNA overexpression using RNA in situ hybridization (RNAscope) and NanoString assay in fresh or archival tumor samples. Pts received rogaratinib 800 mg po BID continuously. Tumor response and safety were assessed. Results: 74 pts with UC were treated with rogaratinib; 73.0% were male, median age was 66 years (range 45-85), and 93.2% had stage IV disease. Rogaratinib was well tolerated, with adverse events being mostly mild or moderate. The most common treatment-emergent adverse events (TEAEs) are shown in the Table. The most common drug-related TEAEs (any grade) were diarrhea (52.7%), increased blood phosphorus (41.9%), and decreased appetite and dry mouth (31.1% each). No ocular toxicities were reported. Increased blood creatinine and acute kidney injury (AKI), regardless of relatedness, were reported in 16.2% and 2.7% of pts, respectively; 1 case of AKI was confirmed as acute tubular necrosis. Of 72 evaluable pts, 15 (20.8%) achieved an objective response; complete and partial responses were observed in 1 (1.4%) and 14 (19.4%) pts, respectively. Stable disease was achieved by 34 pts (47.2%), with a disease control rate of 68.1%. Conclusions: Rogaratinib demonstrated a favorable safety and efficacy profile in pts with tumor FGFR1-3 mRNA-positive UC. TEAEs observed in >25% of pts. Clinical trial information: NCT01976741. [Table: see text]

489 Background: Aberrant activation of fibroblast growth receptor (FGFR) signaling plays a role in UC. Rogaratinib, a pan-FGFR1-4 inhibitor, has promising efficacy and safety in pts with advanced muscle-invasive UC, selected based on FGFR1-3 mRNA overexpression and/or FGFR3-activating mutations/translocations. This Phase II/III, randomized, open-label study evaluated the efficacy of rogaratinib vs CT in pts with FGFR-positive advanced or metastatic UC who received prior platinum CT. We present an ORR analysis for rogaratinib vs CT. Methods: FGFR1/3 mRNA was tested by in situ hybridization of archival tissue. Pts were randomized 1:1 to 800 mg rogaratinib p.o. BID continuously or CT Q3W (i.v., docetaxel 75 mg/m2, paclitaxel 175 mg/m2, or vinflunine 320 mg/m2), and stratified by PIK3CA/ RAS-activating mutations, prior immunotherapy, and modified 4-factor Bellmunt risk score. Results: 87 pts were assigned to rogaratinib and 88 to CT. Overall, 82.9% were male, median age was 69.0 yrs (range: 36-89), 96.6% had stage IV disease, and 2.3% were stage IIIB. PIK3CA/ RAS-activating mutations were present in 11.4% of pts. ORRs of 19.5% and 19.3% (1-sided p=0.56) and disease control rates of 49.4% and 55.7% (p=0.84) were observed for rogaratinib and CT, respectively; median progression-free survival was 2.7 months (95% CI 1.6, 4.2) and 2.9 months (95% CI 2.6, 4.2). Grade 3-4 treatment-emergent adverse events occurred in 40/86 pts (47%) treated with rogaratinib and 46/82 pts (56%) with CT, most commonly anemia (3% vs 15%), neutropenia (1% vs 17%), asthenia (9% vs 1%), lipase increase (8% vs 2%), fatigue (2% vs 6%), and urinary tract infection (2% vs 6%). Exploratory analysis of pts with FGFR3 DNA alterations (4 spot mutations and fusions) showed ORRs of 52.4% with rogaratinib and 26.7% with CT. Conclusions: In pts with FGFR1/3 tumor mRNA-positive UC, rogaratinib had efficacy comparable with standard CT and an acceptable safety profile. Subgroup analysis suggests rogaratinib may be more active in pts with an FGFR3 DNA alteration. Sensitivity analysis of biomarker subgroups is ongoing. Clinical trial information: NCT03410693.

11056 Background: We previously reported that secondary resection (2Surg) improved local relapse free (LRFS) but not overall survival (OS) in a retrospective series of pts with localized STS after unplanned 1st excision. Here we investigated the impact of 2Surg specifically after a first R1 or R2 resection in the 10931 pts with STS of the limb or trunk wall included in the nationwide NETSARC database from 2010 to 2017. Methods: NETSARC (netsarc.org) is a network of 26 reference sarcoma centers with specialized multidisciplinary tumor boards (MDT), funded by the French NCI (INCa). Since 2010, presentation to an MDTB and second pathological review are mandatory for sarcoma pts. Statistics were performed with SPSS23.0. LRFS, metastasis-free survival (MFS), OS compared with the logrank test. Results: The series included 5598 (51.2%) males. Median age was 56.7. Tumor sites: 5295 (49.4%) lower limb, 3670 (33.6%) trunk wall. 1966 (18.0%) upper limb. As previously reported in the entire series, local RFS (LRFS) and RFS (p

11062 Background: EHE is a rare vascular mesenchymal tumor for which there is currently no standard for treatment particularly for metastatic disease. EHE often present metastatic evolution but metastases are not a poor prognostic factor. The aim of this study was to improve knowledge of outcome of EHE patients and see the impact of active surveillance on outcome for metastatic EHE patients. Methods: Patients with EHE treated at three centers in France were included in this retrospective cohort. Univariate analysis of prognostic factors was performed using the Cox model. Survival was estimated using the Kaplan-Meier method and long rank analysis. Results: Fifty-seven patients with EHE were collected in this analysis: 27 (47%) women and 30 (53%) men, with a median age at diagnosis of 39 years (range, 12-83). At diagnosis, 17 (29.8%) patients had a localized tumor, while 40 (70.2%) patients had synchronous metastases. The most commonly affected organs were liver (29.8%), bones (14.0%), skin, lungs and soft tissues (10.5% each). For the 17 patients with localized EHE, the median distant recurrence-free survival after resection of primary was 64.6 months, 95% CI [29.4, NA], (median follow-up of 62.7 months, range, 12.5-234.8). For the 40 patients with metastatic EHE, the median progression-free survival (PFS) was 59.0 months, 95% CI [21.3, NA], (median follow-up of 121.1 months, range, 1.0-202.0). No prognostic factor was identified for localized EHE. For metastatic EHE, age was associated with progression (p = 0.019), and presence of pleural/ascites/pericarditis effusion adversely affected overall survival (OS) (p = 0.002). An initial “wait and see” attitude was proposed to 23 patients (57.5%) while 17 patients (42.5%) were treated at diagnosis with local or systemic treatment (monotherapy or combination of chemotherapy with anthracyclin, taxane, cyclophosphamide). OS were similar in both groups of patients, 174months and 121months for chemotherapy treated patients and active surveillance patients respectively (p = 0.56). Conclusions: Presence of effusion was a significant poor prognostic factor in metastatic EHE patients. Active surveillance could be proposed for asymptomatic patients without effusion but this strategy need to be confirmed in largest or prospective randomized trials.

11021 Background: After we demonstrated the activity of REG in pts with advanced non-adipocytic STS (MirTLO 2016), we conducted a dedicated study in pts previously treated with PAZ+chemo. Methods: We report here the 5th cohort of a double-blind randomized phase 2 trial (NCT01900743). Pts were treated with regorafenib (160mg/d, 21/28d) or placebo (PB). Pts receiving placebo were offered optional cross-over in case of centrally confirmed disease progression. The primary endpoint was centrally-reviewed RECIST-based progression-free survival (PFS), evaluated on the intent-to-treat dataset. A total of 24 events was required to ensure a 90%-power for HR = 0.33 (median PFS, 3·6 vs 1·2 months), with a 1-sided α = 0·1. Overall survival (OS) was a secondary endpoint. Results: From 12/2015 to 10/2017, 37 pts were randomized (18 REG vs 19 PB) and included in the final analysis. The median age was 60 (36-76). There were 28 women (76%). All pts had a performance status 0 or 1. Histological subtypes included 24 leiomyosarcoma (11 vs 13, in REG and PB, respectively), 1 synovial sarcoma (REG), 12 other sarcoma (7 vs 5). All pts had previously been treated with PAZ +chemo (including doxorubicin: 19 vs 17; ifosfamide: 11 vs 3; trabectedin: 11 vs 9; and dacarbazine: 7 vs 6), with 2-6 prior lines. The median relative dose intensity of REG was 0·86, range 0·41-1. Out of 19 pts assigned to placebo, 13 switched to REG after progression. There was no reported objective response. We observed a significant benefit of REG compared to PB in terms of PFS (HR = 0·38; 95%CI, 0·19-0·76; p = 0·007; median PFS = 2·1 vs 1·1 months, respectively), and OS despite the cross-over (HR = 0·41; 95%CI, 0·17-0·98; p = 0·04; median OS = 18·6 vs 8·2 months). Before cross-over, the most common clinically significant grade 3 or higher adverse events were diarrhea (4 vs 0), dyspnea (3 vs 1), arterial hypertension (2 vs 0), hand-foot skin reaction (2 vs 0). Conclusions: The present study demonstrates that regorafenib has a clinically meaningful anti-tumor activity in pts with non-adipocytic soft tissue sarcoma pretreated by both chemotherapy and pazopanib, improving PFS and OS. Clinical trial information: NCT01900743.

Purpose Aberrant mammalian target of rapamycin (mTOR) signaling is common in sarcomas and other malignancies. Drug resistance and toxicities often limit benefits of systemic chemotherapy used to treat metastatic sarcomas. This large randomized placebo-controlled phase III trial evaluated the mTOR inhibitor ridaforolimus to assess maintenance of disease control in advanced sarcomas. Patients and Methods Patients with metastatic soft tissue or bone sarcomas who achieved objective response or stable disease with prior chemotherapy were randomly assigned to receive ridaforolimus 40 mg or placebo once per day for 5 days every week. Primary end point was progression-free survival (PFS); secondary end points included overall survival (OS), best target lesion response, safety, and tolerability. Results A total of 711 patients were enrolled, and 702 received blinded study drug. Ridaforolimus treatment led to a modest, although significant, improvement in PFS per independent review compared with placebo (hazard ratio [HR], 0.72; 95% CI, 0.61 to 0.85; P = .001; median PFS, 17.7 v 14.6 weeks). Ridaforolimus induced a mean 1.3% decrease in target lesion size versus a 10.3% increase with placebo (P < .001). Median OS with ridaforolimus was 90.6 weeks versus 85.3 weeks with placebo (HR, 0.93; 95% CI, 0.78 to 1.12; P = .46). Adverse events (AEs) more common with ridaforolimus included stomatitis, infections, fatigue, thrombocytopenia, noninfectious pneumonitis, hyperglycemia, and rash. Grade ≥ 3 AEs were more common with ridaforolimus than placebo (64.1% v 25.6%). Conclusion Ridaforolimus delayed tumor progression to a small statistically significant degree in patients with metastatic sarcoma who experienced benefit with prior chemotherapy. Toxicities were observed with ridaforolimus, as expected with mTOR inhibition. These data provide a foundation on which to further improve control of sarcomas.

11062 Background: Extraskeletal myxoid chondrosarcoma (EMC) is an exceedingly rare sarcoma, marked by a specific translocation involving the gene NR4A3 that can be rearranged with different partners. Preliminary retrospective data suggest that sunitinib is active, but no formal prospective studies are available. We report on a multicentric European prospective, investigator-driven, Phase 2 study on pazopanib (P) in NR4A3+ advanced EMC patients (pts), carried out by the Spanish, Italian and French Sarcoma groups. Methods: From June 2014 to November 2016, 24 advanced EMC pts entered this study (median age: 64 yrs - disease extent: metastatic 77%, locally advanced 23% - prior medical treatment: 18 (86%) naive; 2 (9%) 1 line, 1 (5%) > 1 line). Path diagnosis and NR4A3 rearrangement (FISH and/or real-time PCR analysis) were centrally confirmed. Pts received P 800 mg/day (relative dose intensity = 0,82%, 658 mg/day), until progression or toxicity. The primary study end-point was response rate as per RECIST 1.1. Secondary end-points were overall survival, progression-free survival (PFS), clinical benefit rate (CBR) (RECIST CR+PR+SD≥6mos). An exploratory evaluation of the correlation between the rearrangement subtype and the outcome is ongoing. Results: 20/24 pts were evaluable for response (1 early death; 3 too early). One patient (5%) had a partial response, 17 (75%) stable disease, 2 (10%) progression as their best RECIST responses. At the time of this analysis, 12 pts were still under treatment, while 12 interrupted P (10 progression, 1 toxicity, 1 other). At a 13-month median follow-up, the median PFS was 13 months (range 1.6-25.1), with 29% pts progression-free at 18 months and a 65% CBR. Median OS was not reached. Conclusions: This Phase 2 study is formally negative since the target of at least 3/21 RECIST responses was not reached. However, looking at PFS, P was associated with a prolonged disease stabilization in a significant proportion of pts. This suggests to further explore the use of P in EMC. Clinical trial information: NCT02066285.

TPS11587Background: PD-L1 is expressed in soft tissue sarcomas (STS) but PD-1 inhibition alone has limited activity. This primary resistance may partly be explained by IDO pathway activation. Epige...

11564Background: Soft tissue sarcomas (STS) are rare tumors accounting for less than 1% of human cancers. While the highest incidence of sarcomas is observed in elderly, this population is often ex...

11508Background: With the exception of a study in translocation-related STS (Kawai, 2015), T has never been compared to BSC in a study including patients (pts) with all sarcoma histotypes. The effi...

11568Background: Retroperitoneal sarcoma (RPS) are complex to treat. We previously reported that management and surgery in a reference center are independent prognostic factors for relapse free sur...

8506Background: The antitumor activity of pembrolizumab, an IgG4 anti-PD-1 monoclonal antibody, was evaluated in patients (pts) with SCLC in KEYNOTE-158 (NCT02628067), a phase 2 basket study of 11 ...

e13593Background: Sarcomas are a group of heterogeneous and rare malignant neoplasms occurring in any anatomical site characterized by various histological subtypes. Because survival is rarely desc...

3059Background: R (anti-VEGFR2) and P (anti-PD-1) are active in patients (pts) with previously-treated non-small cell lung carcinoma (NSCLC), gastric or gastroesophageal adenocarcinoma (G/GEJ) and ...

11523Background: the benefit of neo-adjuvant/adjuvant chemotherapy (CT) in rare bone sarcoma is poorly documented. Methods: retrospective study from the French sarcoma network for bone tumors ResOs...

e23570Background: Endoglin (CD105) is an essential angiogenic receptor expressed on Angiosarcoma (AS) tumor cells and vessels that is upregulated following hypoxia and promotes resistance to VEGF i...

11548Background: This study describes clinical characteristics of a large cohort of aSTS patients (pts) and effectiveness of chemotherapy (CT) based on real world data. Methods: Multicenter retrosp...

11501Background: DT are a group of locally aggressive tumors of fibroblastic origin that can lead to significant morbidity. No randomized trial assessing systemic treatment activity in this rare di...

11504Background: Oral multikinase inhibitor REG has shown activity in GIST and non-adipocytic soft tissue sarcomas. We designed REGOBONE as a non-comparative phase II, double-blind, PL-controlled t...

101 Background: Preclinical evidence suggests simultaneous blockade of vascular endothelial growth factor receptor 2 (VEGFR-2) and programed cell death 1 protein (PD-1) induces synergistic antitumor effects. We assessed safety and preliminary efficacy of ramucirumab (anti-VEGFR-2) plus pembrolizumab (anti-PD-1) in patients (pts) with treatment naïve advanced G/GEJ adenocarcinoma. Methods: This ongoing, multi-cohort, phase 1a/b trial enrolled ECOG PS 0-1 pts with treatment naïve advanced G/GEJ adenocarcinoma with measurable disease and baseline tumor tissue. PD-L1 status was assessed by DAKO PD-L1 22C3 IHC pharmDx assay using the combined positive score with ≥1% being positive. Ramucirumab was administered at 8 mg/kg on Days 1 & 8 with pembrolizumab 200 mg on Day 1 q3W. Primary objective was safety and tolerability of study treatment; preliminary efficacy was examined. Results: As of 31-July-2017, 28 treatment naïve G/GEJ adenocarcinoma pts were treated. Median age was 63 y, 75% male, 57% had ECOG PS of 0, and 68% were PD-L1 positive. At data cutoff, 8 (28%) pts continued to receive study treatment. Median duration of follow-up was 8.1 mo (IQR 6-10). Median treatment duration was 4.3 mo (IQR 2-7). All grades treatment-related adverse events (TRAEs) occurred in 27 (96%) of patients; TRAEs in ≥15% of pts were fatigue (36%), hypertension (25%) and headache (18%). Seventeen (61%) pts experience grade 3 TRAEs, most commonly hypertension (14%), diarrhea (11%), and elevated alanine (7%) or aspartate (7%) aminotransferase. No grade 4-5 TRAEs occurred. An objective response was achieved by 7 (25%) pts, 6 positive and 1 negative for PD-L1. Disease control rate was 68%. Median time to response was 2.7 mo (95% CI 1.3-2.8) and median duration of response was 10 mo (95% CI 9.7-10.3). Median progression-free survival was 5.3 mo (95% CI 3.2-11). Median overall survival has not been reached. Conclusions: Ramucirumab plus pembrolizumab demonstrated encouraging antitumor activity in treatment naïve advanced G/GEJ adenocarcinoma with no grade 4 TRAE observed. Clinical trial information: NCT02443324.

11053 Background: There is a good rationale for immunotherapy in sarcoma. We report results of the first open-label multicentre phase 2 study assessing the anti-PD-1 antibody pembrolizumab in combination with metronomic cyclophosphamide (CP) in patients (pts) with advanced soft tissue sarcomas (STS) and gastro-intestinal stromal tumor (GIST). Methods: This trial included 4 cohorts of pts with advanced STS: leiomyosarcoma (LMS), undifferentiated pleomorphic sarcoma (UPS), other sarcomas (Others), and GIST. All pts received CP 50 mg BID one week on, one week off, and pembrolizumab 200mg IV q21 days. The primary endpoint encompassed non-progression and objective response at 6 months per RECIST evaluation criteria v1.1 for LMS, UPS, and Others, and 6-month non-progression for GIST. Correlative studies of immune biomarkers were planned on pt’s tumor and plasma samples. Results: Between June 2015 and July 2016, 57 pts were included, and 50 were assessable for efficacy. Three pts experienced tumor shrinkage resulting in a partial response (PR) in one of them. The 6-month non-progression rate was 0%, 0%, 14.3% (95%CI 1.8-42.8), and 11.1% (95%CI 2.8-48.3) in LMS, UPS, Others, and GIST respectively. The most frequent adverse events were grade 1 or 2 fatigue, diarrhea, anemia. The only pt who experienced PR was the only one with a PD-L1-positive staining in more than 10% of immune cells on archived tumor sample. A strong macrophage infiltration was observed in tumor samples, and these macrophages largely expressed the inhibitory enzyme Indoleamine-2,3-dioxygenase-1 (IDO1). Moreover, a significant increase of the kynurenine/tryptophane ratio was observed in pts plasma samples during study treatment (p =0.0007). Conclusions: PD-1 inhibition has limited activity in advanced STS and GIST. This primary resistance may be explained by the low percent of PD-L1 positivity in these tumors, and an immune-suppressive tumor microenvironment resulting from macrophage infiltration and IDO1 pathway activation. Further strategies assessing drugs such as CSF1-R inhibitors and/or IDO inhibitors combined with anti-PD-1/PD-L1 in selected sarcoma subtypes are warranted. Clinical trial information: NCT02406781.

TPS11081 Background: AAS is an aggressive soft tissue sarcoma (STS) of endothelial cell origin with an expected median overall survival of 8-12 months. Pazopanib (P) is approved for treatment of advanced STS following progression on chemotherapy. In a retrospective study of 40 AAS patients treated with single agent P the median PFS was 3.1 months and median OS 9.9 months with no complete responses. Endoglin is an essential angiogenic receptor expressed on AAS that is upregulated following VEGF inhibition, and TRC105, an endoglin antibody, given with P produced durable complete responses in AAS patients with median PFS of 5.6 months in refractory patients including those receiving prior P. The TAPPAS trial is the first randomized Phase 3 trial performed in AAS, and was initiated following protocol assistance from the EMA and Special Protocol Assessment from the FDA. Methods: TAPPAS is a randomized multicenter study of TRC105/P vs P alone in the United States and Europe that is actively enrolling cutaneous and non-cutaneous AAS patients and incorporates an adaptive enrichment design. Key inclusion criteria: 0, 1 or 2 prior lines of therapy, ECOG ≤ 1. Primary endpoint is PFS and secondary endpoints include ORR and OS. The initial sample size of 124 patients, followed until 95 PFS events, provides more than 80% power to detect a hazard ratio of 0.55. At the time of interim analysis, projected to occur upon the occurrence of 40 events in approximately 70 patients, the result will be classified as belonging to either the favorable, promising, enrichment or unfavorable zones, based on conditional power. The sample size and PFS events will be unchanged in the favorable and unfavorable zones, and will be increased to a total of 200 patients followed for 170 PFS events in the promising zone. The trial will enroll 100 additional patients, with cutaneous disease only, in the enrichment zone and will follow them until 110 events are observed in the total cutaneous population. An independent DMC will follow the trial for safety and futility. The adaptive design requires the enrollment of fewer patients, preserves type-1 error, and protects power to detect a clinically meaningful survival benefit. (NCT 02979899). Clinical trial information: NCT02979899.