Funding Institution Research Articles

484P - Activity of afatinib in patients (pts) with EGFR mutation-positive (EGFRm+) NSCLC and baseline brain metastases: Pooled analysis of three large phase IIIb trials

BackgroundIn the LUX-Lung (LL) 3 and 6 trials, first-line afatinib significantly improved PFS vs chemotherapy in pts with EGFRm+ NSCLC and baseline brain mets (HR, 0.50; P = 0.03).1 In LL7, similar PFS improvement with afatinib vs gefitinib was observed in pts with, and without, brain metastases (HR, 0.76 and 0.74; Pint=0.93).2 Competing risk analysis of LL3/6 demonstrated low risk of de novo CNS progression in pts treated with afatinib (6%).3 Here, we assess afatinib in pts with brain metastases treated in a ‘real-world’ setting. MethodsRetrospective pooled analysis of three ‘real-world’ studies: a multi-center expanded-access program in Korea (1200.193); an Asian phase IIIB trial (1200.66); a global phase IIIB trial (mainly Europe; 1200.55). All three studies recruited EGFR TKI-naïve pts with EGFRm+ NSCLC who received afatinib 40 mg/day, including pts with asymptomatic brain metastases. Dose reduction was permitted (minimum 20 mg/day). Endpoints included investigator-assessed progression-free survival (PFS), time to symptomatic progression (TTSP) and objective response rate (ORR). ResultsA total of 1108 pts were treated with afatinib. Baseline characteristics were: median age, 61 yrs; female, 58%; ECOG PS of 0/1/2, 26%/70%/4%; uncommon EGFR mutations: 18%; first-line afatinib, 69%. In total, 213 pts had brain metastases. Median PFS and TTSP in these pts were 10.6 months (95% CI 9.1–12.8) and 13.7 months (95% CI 11.0–14.8), respectively. When restricted to pts harboring common EGFR mutations, median PFS and TTSP were 11.7 months (95% CI 10.1–13.8) and 14.4 months (95% CI 12.9–16.4), respectively. ORR was 57% (59% in pts with common EGFR mutations); median duration of response was 11.1 months (95% CI 8.3–12.3). ConclusionsConsistent with clinical trial data, afatinib is active in pts with EGFRm+ NSCLC and brain metastases treated in a real-world setting in Asian and Caucasian pts. 1. Schuler, M. et al. J Thorac Oncol 2016;11:380–90 2. Park, K. et al. Lancet Oncol 2016;17:577–89 3. Girard, N. Future Oncol. 2018;14:1117–32. Clinical trial identificationNCT01931306; NCT01953913; NCT01853826. Editorial acknowledgementLynn Pritchard of GeoMed, an Ashfield company, part of UDG Healthcare plc. Legal entity responsible for the studyBoehringer Ingelheim. FundingBoehringer Ingelheim. DisclosureF. de Marinis: Honoraria (self): Roche; Honoraria (self): AstraZeneca; Advisory / Consultancy: Takeda; Research grant / Funding (institution): Boehringer Ingelheim. V. Lee: Honoraria (self), Advisory / Consultancy: Pfizer; Honoraria (self): Eli Lilly; Honoraria (self), Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: AstraZeneca; Honoraria (self), Advisory / Consultancy: Merck Sharp & Dohme. L. Clementi: Full / Part-time employment: Boehringer Ingelheim. W. Tang: Full / Part-time employment: Boehringer Ingelheim. D.C-L. Huang: Full / Part-time employment: Boehringer Ingelheim. A. Cseh: Full / Part-time employment: Boehringer Ingelheim. K. Park: Advisory / Consultancy: AMGEN; Advisory / Consultancy: Astellas; Advisory / Consultancy, Speaker Bureau / Expert testimony, Research grant / Funding (self): AstraZeneca; Advisory / Consultancy: BluePrint; Advisory / Consultancy: BMS; Advisory / Consultancy, Speaker Bureau / Expert testimony: Boehringer Ingelheim; Advisory / Consultancy: Eli Lilly; Advisory / Consultancy: Hanmi; Advisory / Consultancy: KHK; Advisory / Consultancy: Loxo; Advisory / Consultancy: Merch KGaA; Advisory / Consultancy, Research grant / Funding (self): MSD; Advisory / Consultancy: ONO; Advisory / Consultancy: Roche. C. Zhou: Honoraria (self): BI; Honoraria (self): Roche; Honoraria (self): Sanofi; Honoraria (self): Hengrui; Honoraria (self): Qilu; Honoraria (self): MSD. Y-L. Wu: Honoraria (self), Advisory / Consultancy: AZ; Advisory / Consultancy, Research grant / Funding (self): Roche; Honoraria (self), Advisory / Consultancy: BMS; Full / Part-time employment: Guangdong Provincial People's Hospital, China; Honoraria (self): Eli Lilly; Honoraria (self): Pfizer; Honoraria (self), Research grant / Funding (self): BI. All other authors have declared no conflicts of interest.

225O - Impact of radiological disease progression on the health status of patients (pts) with BRCA-mutated advanced ovarian cancer (OC) treated with maintenance olaparib (OL) or placebo (PL) after first-line platinum chemotherapy

BackgroundIn SOLO1 (NCT01844986), maintenance OL significantly improved progression-free survival vs PL (HR 0.30; 95% CI 0.23–0.41; Moore et al. NEJM 2018) following response to first-line platinum chemotherapy in pts with BRCA-mutated advanced OC. We compared pt health status pre- and post-radiological progression (RECIST) in SOLO1 using the EQ-5D-5L descriptive system and visual analogue scale (VAS). MethodsEQ-5D-5L data were collected at baseline, day 29, every 12 weeks for 3 years, then every 24 weeks until data cut off (May 2018). Using the last EQ-5D-5L pre-progression as baseline, we assessed changes in each domain score, the proportion reporting any problem (levels 2–5) by domain, and Paretian classification of health change (PCHC) at first post-progression EQ-5D-5L. Impact of progression on VAS was assessed by mixed-effects repeated measures analysis. Data were pooled across study arms. Results74/100 OL pts and 81/95 PL pts with RECIST progression completed the EQ-5D-5L pre- and post-progression. Pre-progression, any problems were reported in 56%, 54%, 40%, 31% and 5% for anxiety/depression, pain/discomfort, usual activity, mobility, and self care respectively. Post-progression, proportions reporting problems with anxiety/ depression (68%), pain/discomfort (65%) and self care (15%) increased, while usual activity (43%) and mobility (30%) were unchanged. Across all levels, worsening was reported in anxiety/depression (34%), pain/discomfort (27%) and self care (13%). By PCHC, 56% reported worsening in at least one domain, with 37% reporting worse overall health status immediately post-progression. Progression was associated with a statistically significant and meaningful reduction in VAS (-0.0714, P < 0.0001). ConclusionsRadiological progression after response to first-line platinum chemotherapy is associated with worsening health status, with anxiety/depression and pain/discomfort most affected. These results highlight the impact of progression on pts and pt-reported health benefits of delaying progression with maintenance OL. Clinical trial identificationNCT01844986. Editorial acknowledgementMedical writing assistance was provided by Catherine Risebro, PhD, from Mudskipper Business Ltd, funded by AstraZeneca. Legal entity responsible for the studyAstraZeneca. FundingAstraZeneca. DisclosureM.L. Friedlander: Honoraria (self), Honoraria (institution), Advisory / Consultancy: AstraZeneca; Non-remunerated activity/ies, Consulting: AbbVie; Research grant / Funding (institution): Beigene; Advisory / Consultancy: Lilly; Advisory / Consultancy: MSD; Advisory / Consultancy: Takeda. R. Hettle: Shareholder / Stockholder / Stock options, Full / Part-time employment: AstraZeneca. E. Parkhomenko: Honoraria (institution), Advisory / Consultancy, Full / Part-time employment, Employee of Parexel, which has received consultancy fees from AstraZeneca: AstraZeneca.

483P - A phase IIIb, open-label study of afatinib in EGFR TKI-naïve patients with EGFR mutation-positive NSCLC: A biomarker analysis

BackgroundTreatment with an EGFR TKI is the standard of care for patients (pts) with EGFR mutation-positive (EGFRm+) NSCLC, based on findings from randomized controlled trials, but acquired resistance is inevitable. A phase IIIb study (NCT01853826) was conducted to assess outcomes with afatinib in a broad EGFRm+ NSCLC patient population, similar to real-world practice. Here we report results from a sub-study of the trial, to assess the relationship between afatinib efficacy and the presence of biomarkers associated with resistance to EGFR TKIs. MethodsEGFR TKI-naïve pts with locally advanced/metastatic EGFRm+ NSCLC and ECOG PS 0–2 received afatinib starting at 40 mg/day. Tumor biopsies were performed prior to study entry; baseline T790M status (presence vs absence), and total BIM, BIMEL and mTOR mRNA expression (high vs low, according to median values) were analyzed for patients who consented to sub-study participation. Exploratory efficacy analyses were conducted. Time to symptomatic progression (TTSP) and progression-free survival (PFS) were calculated using the Kaplan–Meier method. ResultsA total of 113 pts were included in this sub-study (data cut-off 21 June 2018). Baseline characteristics, n (%): white race, 112 (99); female, 70 (62); 1st/2nd/≥3rd-line therapy, 94/16/3 (83/14/3); ECOG PS 0/1, 106 (94); brain metastases, 16 (14); common/uncommon mutations, 105/8 (93/7). Median time on treatment was 581 days. Objective response/disease control rates were 58%/95%. Median (95% CI) TTSP was 20.6 months (18.9–24.9) and PFS was 19.8 months (16.1–23.5). TTSP and PFS by biomarker status at baseline are reported in the table.Table483PTableTTSPPFSMedian (95% CI)HR (95% CI)Median (95% CI)HR (95% CI)T790M*Present (n = 51) Absent (n = 60)19.5 (14.9–25.1) 21.4 (17.7–26.3)1.0 (0.67–1.60)18.8 (14.1–24.0) 21.4 (15.9–26.1)1.1 (0.7–1.7)BIM total mRNA expression†High (n = 35) Low (n = 35)23.7 (14.7–37.1) 25.7 (19.3–35.8)1.2 (0.7–2.1)21.8 (14.1–28.5) 26.0 (18.9–35.5)1.2 (0.7–2.1)BIMEL mRNA expression‡High (n = 22) Low (n = 21)19.1 (11.9–24.8) 24.9 (14.8–42.2)1.3 (0.6–2.7)18.8 (11.6–23.5) 25.2 (16.1–38.6)1.4 (0.7–2.9)mTOR mRNA expression§High (n = 39) Low (n = 37)22.3 (14.8–29.5) 26.1 (19.1–35.8)1.0 (0.6–1.7)21.8 (14.3–29.5) 23.1 (14.5–35.5)1.0 (0.6–1.8)Non-evaluable or missing: *n=2; †n = 43; ‡n = 36; §n = 37. ConclusionsIn this broad population of EGFR TKI-naïve pts with EGFRm+ NSCLC who were treated with afatinib, biomarker analysis indicated no notable differences in efficacy outcomes by baseline T790M status (presence vs absence), or baseline mRNA expression of BIM, BIMEL or mTOR (high vs low). Clinical trial identificationNCT01853826. Editorial acknowledgementLaura Winton of GeoMed, an Ashfield company, part of UDG Healthcare plc. Legal entity responsible for the studyBoehringer Ingelheim. FundingBoehringer Ingelheim. DisclosureM. Hochmair: Advisory / Consultancy, Speaker Bureau / Expert testimony: AstraZeneca; Advisory / Consultancy, Speaker Bureau / Expert testimony: BMS; Advisory / Consultancy, Speaker Bureau / Expert testimony: Boehringer Ingelheim; Advisory / Consultancy, Speaker Bureau / Expert testimony: Merck Sharp & Dohme, ; Advisory / Consultancy: Novartis; Advisory / Consultancy, Speaker Bureau / Expert testimony: Roche; Advisory / Consultancy: Takeda; Speaker Bureau / Expert testimony: Pfizer. R. Ramlau: Advisory / Consultancy: MSD; Advisory / Consultancy, Travel / Accommodation / Expenses: BMS; Advisory / Consultancy, Travel / Accommodation / Expenses: Roche; Advisory / Consultancy, Travel / Accommodation / Expenses: Boehringer Ingelheim; Advisory / Consultancy, Travel / Accommodation / Expenses: AstraZeneca. J. Skrickova: Advisory / Consultancy: Boehringer Ingelheim; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Roche. P. Bidoli: Advisory / Consultancy: Eli Lilly; Advisory / Consultancy: BMS ; Advisory / Consultancy: Boehringer Ingelheim. L. Clementi: Full / Part-time employment: Boehringer Ingelheim. A. Cseh: Full / Part-time employment: Boehringer Ingelheim. F. De Marinis: Honoraria (self): Roche; Honoraria (self): AstraZeneca; Advisory / Consultancy: Takeda; Research grant / Funding (institution): Boehringer Ingelheim. All other authors have declared no conflicts of interest.

230P - Olaparib maintenance therapy in patients (pts) with a BRCA1 and/or BRCA2 mutation (BRCAm) and newly diagnosed advanced ovarian cancer (OC): SOLO1 China cohort

BackgroundSOLO1 (NCT01844986) is a randomized, double-blind, phase III trial evaluating the efficacy and safety of the PARP inhibitor, olaparib, as maintenance monotherapy in newly diagnosed advanced OC pts with a BRCAm. A separate pt cohort evaluated the efficacy and safety of olaparib in Chinese pts in this setting. MethodsThe China cohort of SOLO1 planned to enrol ∼53 newly diagnosed OC pts who had completed first-line platinum-based chemotherapy and were in clinical complete or partial response. This sample size provided around a 90% chance to observe a hazard ratio (HR) of < 1, assuming a true HR of 0.62. Pts were randomized 2:1 to olaparib (300 mg bid; tablet) vs placebo. The primary endpoint was investigator-assessed progression-free survival (PFS; modified RECIST v1.1). Sensitivity analysis of PFS was performed by blinded independent central review (BICR). ResultsAll 64 randomized pts received study treatment (olaparib, n = 44; placebo n = 20). Median follow-up was ∼30 months in both arms. Median PFS was not reached in the olaparib arm and was 9.3 months in the placebo arm (Table).Table230PUnlabelled ImagePFS events, nMedian PFS, monthsHR (95% CI; P value)Olaparib (n = 44)Placebo (n = 20)Olaparib (n = 44)Placebo (n = 20)Full analysis set (n = 64)PFS, investigator assessed (48.4% maturity)1813NR9.30.46 (0.23, 0.97; 0.0320)PFS, BICR (39.1% maturity)1312NR9.30.39 (0.17, 0.86; 0.0168)CI, confidence interval; NR, not reached.The most common AEs in the olaparib group were nausea (n = 28, 63.6%), anaemia (n = 25, 56.8%) and vomiting (n = 18, 40.9%). Grade ≥3 AEs occurred in 56.8% of olaparib pts vs 30.0% of placebo pts; the most common grade ≥3 AE was anaemia (n = 16, 36.4%). Olaparib dose interruptions, reductions and discontinuations occurred in 56.8%, 27.3% and 6.8% of pts, respectively (vs 30.0%, 10% and 0% of pts in the placebo arm). ConclusionsIn the China cohort of SOLO1, a clinically relevant improvement in investigator-assessed PFS was observed in newly diagnosed OC pts receiving olaparib maintenance therapy. Olaparib treatment led to a 54% reduction in risk of progression or death vs placebo. The safety results were consistent with the known profile of olaparib in Chinese pts. Clinical trial identificationNCT01844986, release date 19 February 2016. Editorial acknowledgementMedical writing assistance was provided by Laura Smart, MChem, from Mudskipper Business Ltd, funded by AstraZeneca and Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA (MSD). Legal entity responsible for the studyAstraZeneca. FundingAstraZeneca. DisclosureJ. Liu: Research grant / Funding (institution): AstraZeneca. X. Ma: Full / Part-time employment: AstraZeneca. J. Zhang: Shareholder / Stockholder / Stock options, Full / Part-time employment: AstraZeneca. All other authors have declared no conflicts of interest.

73P - Cabozantinib in combination with anti-PD1 immune checkpoint inhibitor in syngeneic tumour mouse models

BackgroundCabozantinib (C), a tyrosine kinase inhibitor against VEGFR2, RET, MET and Axl, has demonstrated clinical activity in RCC and 2L HCC, and has been shown to promote an immune-permissive environment. C is being investigated in clinical trials in combination with immune checkpoint inhibitors (ICI). This preclinical study aimed to determine systemic cytokine profile changes induced by the combination of C + αPD1 and assess cooperativity in anti-tumoral effects. MethodsSyngeneic colon (CT26, Colon38) and bladder (MBT2) tumour models were established in BALB/C, C57Bl6 or C3H/HEJ mice. C (10 or 30 mg/kg po qd) and αPD1 (10 mg/kg ip 2x weekly) were administered either simultaneously for up to 30 days or with a 14-day delay for αPD1, and serum was collected. ResultsAfter 30 days' coadministration of C (30 mg/kg) and αPD1, serum cytokines and chemokines (VEGFA, IL3, GM-CSF, MIP1beta, IL17, CCL5, IFNγ, IL10, IL5) were elevated with a median of 72x (range 39–1207x), as compared to 30-days' single administration of αPD1 or C with a median of 3.7x (1.4–135x) and 1.0x (0.5–1.9x), respectively, relative to vehicle-treated animals. Delayed treatment with αPD1 did not result in the same increase in analyte levels as coadministration (median of 1.74x [1–9.9x]). For anti-tumour efficacies, two C doses were explored in CT26 (T/C10/30 67/38% - tumour growth inhibition T/C ratios at 10 or 30 mg/kg), MBT2 (T/C10/30 37/14%) and Colon38 (T/C10/30 21/12%) models. C at 30 mg/kg led to strong anti-tumour efficacy in MBT2 and Colon38, limiting any ICI combinatorial impact to the post-treatment tumour expansion. For CT26 (30 mg/kg) and MBT2 (10 mg/kg), simultaneous dosing of C and αPD1 showed greater anti-tumour effects than single agents despite only limited effects of αPD1 alone. C + αPD1 more than doubled the treatment-specific time that tumours required to reach 1000mm3 compared to C alone. ConclusionsThe results indicate that simultaneous treatment of C + αPD1 leads to strong systemic increases in key cytokines and chemokines and can result in robust anti-tumour efficacy. Our results support the clinical exploration of C + ICI combinations and the potential of this TKI for use in further clinical settings. Editorial acknowledgementIsabelle Kaufmann, PhD of Oxford PharmaGenesis, Oxford, UK for provided medical writing support, which was sponsored by Ipsen, Abingdon, UK. Legal entity responsible for the studyIpsen. FundingIpsen. DisclosureR. Sparks: Full / Part-time employment: Ipsen. S. Rolland: Full / Part-time employment: Ipsen. S. Klinz: Full / Part-time employment: Ipsen. F. Meyer-Losic: Full / Part-time employment: Ipsen. J. Nakhle: Full / Part-time employment: Ipsen. R. Delille: Full / Part-time employment: Ipsen. M. Hillairet de Boisferonc: Research grant / Funding (institution), Contractual obligation: Ipsen.

482P - Interim analysis from a phase IIIb, open-label study of afatinib in EGFR TKI-naïve patients (pts) with EGFR mutation-positive (EGFRm+) NSCLC

BackgroundFirst-line afatinib significantly improved median progression-free survival (PFS) in pts with EGFRm+ NSCLC vs chemotherapy (CT) in LUX-Lung (LL) 3/6 (HR [95% CI]: 0.58 [0.43, 0.78]/0.28 [0.20, 0.39]), and vs gefitinib in LL7 (0.73 [0.57, 0.95]). We conducted a phase IIIb study of afatinib in a broader patient population, similar to real-world practice, of treatment-naïve or CT pre-treated pts with EGFRm+ NSCLC. Here we present an interim analysis. MethodsEGFR TKI-naïve pts with locally advanced/metastatic EGFRm+ NSCLC and ECOG PS 0–2 received afatinib 40 mg/day; dose reduction was permitted (min. 20 mg/day). Primary endpoint: adverse events (AEs; descriptive fashion). Efficacy was assessed and post-hoc subgroup analysis conducted. Results479 pts were included (data cut-off: 30/04/18): Caucasian/Asian: 97%/2%; female: 66%; 1st/2nd/≥3rd-line therapy: 78%/17%/5%; ECOG PS 0 or 1/2: 92%/8%; brain metastases: 17%; common/uncommon [C/U] mutations: 87%/13%. Median time on afatinib: 359 days. Objective response rate: 46%. Disease control rate: 86%. Median time to symptomatic progression (TTSP) and PFS were 14.9 (95% CI: 13.8, 17.6) and 13.4 (11.8, 14.5); subgroup analysis is in the table. Most common grade ≥3 afatinib-related AEs were diarrhea (16%) and rash (11%). AEs led to dose reduction in 258 (54%) pts (most frequently diarrhea 25%, rash 11%) and to afatinib discontinuation in 105 (22%) pts (malignant neoplasm progression 3%, diarrhea 3% [rash 0.8%]). Afatinib-related serious AEs occurred in 39 (8%) pts.Table482PTableTTSP, mosPFS, mosBaseline characteristic (n)Median95% CIMedian95% CILine of tx1 (374)15.614.1, 18.513.812.6, 15.22 (81)14.711.3, 20.613.28.3, 17.7≥3 (24)8.13.7, 14.46.63.2, 12.6MutationsaDel 19b (232)19.315.6, 21.815.913.9, 19.1L858Rb (162)14.512.7, 17.913.111.5, 15.2Uc (84)7.45.7, 9.06.04.2, 8.1ECOG PS, inc. mutations0–1 (442)a15.814.4, 18.813.812.8, 15.2Cb (364)18.215.5, 19.815.213.8, 17.7U (77)7.45.7, 9.76.64.6, 8.22 (36)8.95.7, 13.26.22.5, 11.6Cb (30)8.95.7, 13.97.73.9, 13.2U (6)7.00.9, 13.01.40.4, 6.0Brain metsaNo (395)15.814.1, 18.813.912.7, 15.5Yes (83)13.79.7, 17.210.18.2, 13.9aMissing n = 1.bDel 19/L858R only.cIncludes, n (%): ex 20 ins: 37 (44). ConclusionsThis analysis indicates a predictable and manageable safety profile for afatinib, consistent with the pivotal LL trials, and encouraging efficacy findings in this broad patient population. As expected, TTSP/PFS were longer for pts with ECOG PS 0/1 vs 2, and for pts with Del 19 or L858R mutations, vs those with uncommon mutations, which may be due to the high prevalence (44%) of pts with exon 20 insertions included in this study. Clinical trial identificationNCT01853826. Editorial acknowledgementChristina Jennings of GeoMed, an Ashfield company, part of UDG Healthcare plc. Legal entity responsible for the studyBoehringer Ingelheim. FundingBoehringer Ingelheim. DisclosureF. De Marinis: Honoraria (self): Roche; Honoraria (self): AstraZeneca; Advisory / Consultancy: Takeda; Research grant / Funding (institution): Boehringer Ingelheim. M. Hochmair: Advisory / Consultancy, Speaker Bureau / Expert testimony: Boehringer Ingelheim; Advisory / Consultancy, Speaker Bureau / Expert testimony: Merck Sharp & Dohme; Advisory / Consultancy: Novartis; Advisory / Consultancy, Speaker Bureau / Expert testimony: Roche; Speaker Bureau / Expert testimony: AstraZeneca; Speaker Bureau / Expert testimony: Bristol-Myers Squibb; Speaker Bureau / Expert testimony: Pfizer. M.R. Migliorino: Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Travel / Accommodation / Expenses: B.I.; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Travel / Accommodation / Expenses: ROCHE; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Travel / Accommodation / Expenses: MSD; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Travel / Accommodation / Expenses: AstraZeneca; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony: PFIZER. G.Z. Mukhametshina: Advisory / Consultancy: Association of oncologists of Russian Federation; Full / Part-time employment: State Autonomous Healthcare Institution «Republican Clinical Oncology Dispensary of the Ministry of Health of the Republic of Tatarstan». M. Schumacher: Honoraria (self), Advisory / Consultancy: BMS; Honoraria (self), Advisory / Consultancy: MSD; Honoraria (self), Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: Pfizer; Honoraria (self), Advisory / Consultancy: AstraZeneca. S. Novello: Advisory / Consultancy, Speaker Bureau / Expert testimony: AstraZeneca; Advisory / Consultancy, Speaker Bureau / Expert testimony: BI; Advisory / Consultancy, Speaker Bureau / Expert testimony: MSD; Advisory / Consultancy, Speaker Bureau / Expert testimony: BMS; Advisory / Consultancy, Speaker Bureau / Expert testimony: Roche; Advisory / Consultancy, Speaker Bureau / Expert testimony: AbbVie; Advisory / Consultancy, Speaker Bureau / Expert testimony: Celgene; Advisory / Consultancy, Speaker Bureau / Expert testimony: Eli Lilly; Advisory / Consultancy, Speaker Bureau / Expert testimony: Pfizer; Advisory / Consultancy, Speaker Bureau / Expert testimony: Takeda. R. Dziadziuszko: Honoraria (self): Pfizer; Honoraria (self): Boehringer Ingelheim; Honoraria (self), Travel / Accommodation / Expenses: AstraZeneca; Honoraria (self), Travel / Accommodation / Expenses: Roche; Honoraria (self): MSD; Honoraria (self): Bristol-Myers Squibb. W. Tang: Full / Part-time employment: Boehringer Ingelheim. L. Clementi: Full / Part-time employment: Boehringer Ingelheim Italia SpA. A. Cseh: Full / Part-time employment: Boehringer Ingelheim. All other authors have declared no conflicts of interest.

144P - Severe hypovitaminosis D in metastatic gastric cancer patients from the Northern and Southern hemispheres: Data from the EXPAND phase III trial

BackgroundOur primary goal was to evaluate the association between vitamin D (25-OHD) plasma levels and overall survival (OS) in metastatic gastric cancer patients receiving first-line chemotherapy in the international phase III EXPAND study (EudraCT 2007-004219-75). Here we report on 25-OHD plasma levels and OS according to geographic hemisphere of residence. Methods25-OHD pre-therapeutic plasma levels were measured in 630 patients using chemiluminescent immunoassay (Abbott Laboratories, Illinois). All patients were treated with chemotherapy based on capecitabine and cisplatin ± cetuximab. The Cox proportional hazard regression model was used to analyse the association between 25-OHD and survival in both treatment arms. Results549 and 73 patients were recruited in the Northern and Southern hemispheres, respectively. The majority had severe hypovitaminosis D at baseline prior to start of treatment. Only 9.2% of patients had a normal 25-OHD plasma levels (applying accepted criteria of 25-OHD insufficiency with plasma levels < 75 nmol/l; deficiency defined as < 50 nmol/l). Median 25-OHD plasma level was 38.2 nmol/l (IQR 25.8; 58.9) and mean was 43.5 nmol/l. Patients from the Southern versus Northern hemisphere had slightly higher 25-OHD levels (median 45.6 vs. 37.4 nmol/l; p = 0.0172; mean 48.92 vs. 42.74; p = 0.03 ). This difference persists when adjusting for season in a linear regression model. As expected, we observed the highest levels of 25-OHD in samples drawn during the summer season and the lowest values in winter. No proof for a negative impact of hypovitaminosis D on survival or treatment response was detected and no difference in OS was seen between patients from the Southern vs. Northern hemisphere (median 8.3 months [95%CI 7.4-10.6] vs. 10.5 [95%CI 9.4-11.3]; Hazard Ratio = 1.195; p = 0.2001). ConclusionsNo prognostic value of vitamin D deficiency was found in the EXPAND cohort. However, the majority of patients with metastatic gastric cancer have a severe vitamin D deficiency with only minor differences comparing Southern versus Northern hemisphere of residency. Legal entity responsible for the studyMasaryk Memorial Cancer Institute. FundingMasaryk Memorial Cancer Institute, University Leipzig. DisclosureR. Obermannova: Advisory / Consultancy, Speaker Bureau / Expert testimony: Servier; Advisory / Consultancy, Research grant / Funding (self): Merck; Advisory / Consultancy, Speaker Bureau / Expert testimony: Eli Lilly. F. Lordick: Advisory / Consultancy, Speaker Bureau / Expert testimony: Amgen, AstraZeneca, Astellas, Biontech, BMS, Eli Lilly, Elsevier, Excerpta Medica, Imedex, Infomedica, Iomedico AG, Medscape, MedUpdate GmbH, Merck Sharp Dohme, Merck Serono, Oncovis Gmbh, Promedicis, Springer Nature Group, StreamedUp!, Zymeworks; Research grant / Funding (institution): BMS. All other authors have declared no conflicts of interest.

290O - Long-term results of phase II trial of reduced modified clinical target volume in low-risk nasopharyngeal carcinoma treated with intensity modulated radiotherapy

BackgroundIntensity modulated radiation therapy (IMRT) is the standard of care for nasopharyngeal carcinoma (NPC). Part of the IMRT quality assurance process involves accurate target volume delineation, which is crucial for local control and survival. Current practice utilizes a uniform contouring method for clinical target volume (CTV) for all stages of NPC. Here, we conducted a single-arm phase 2 trial investigating the tumor control rates of reduced CTV margins and corresponding doses in low-risk, early-stage NPC. MethodsPatients with biopsy-proven stage I-IIb (6th UICC/AJCC) NPC were enrolled. All patients were treated with IMRT alone. Two CTVs (CTV1 [high risk] and CTV2 [low risk]) were outlined; CTV1 was defined by grossly identified tumour (on MRI or CT) plus 5-mm margin (3-mm posteriorly); CTV2 was CTV1 plus 5-mm margin (3-mm posteriorly). 60 Gy and 50–54 Gy in 30 fractions were prescribed to CTV1 and CTV2, respectively. Primary end-point was locoregional recurrence free survival (LRRFS). ResultsFrom May 2001 to August 2006, 103 patients were recruited; all patients completed IMRT as planned (time duration 39-61 days). At a median follow-up of 12.6 years, five patients developed locoregional failures; of which one was in-field, and four were regional recurrence (two in-field and two marginal). 10-year LRRFS, distant metastasis free survival (DMFS), disease specific survival (DSS) and overall survival (OS) were 95.0% and 94.1%, 94.0% and 91.1%, respectively. The most common grade 1-2 late toxicities included subcutaneous fibrosis (79.6%), hearing loss (53.4%) and skin dystrophy (43.2%); grade 3 included subcutaneous fibrosis (2.9%) and hearing loss (2.9%); and no grade 4 late toxicity. ConclusionsReduced CTV margins and corresponding doses results in optimal longterm tumour control, with minimal late adverse events. Clinical trial identificationNCT03839602. Legal entity responsible for the studyProf. Chong Zhao. FundingHas not received any funding. DisclosureM.L.K. Chua: Honoraria (self), Advisory / Consultancy: Janssen, Astellas, Varian, Ferring Singapore and AstraZeneca; Research grant / Funding (institution): Ferring Singapore, GenomeDx Biosciences, Varian, MedLever. All other authors have declared no conflicts of interest.

254P - Cyclin E1 is a shared biomarker of subsets of high grade serous ovarian cancer (HGSOC) and basal like breast cancer (BLBC)

BackgroundHGSOC and BLBC are aggressive cancers that share aetiology, co-occur frequently and share molecular signatures including cyclin E1 deregulation. In HGSOC, we have described two cyclin E1 high groups, with and without associated gene amplification. We aimed to determine which subsets of these disease have the greatest similarity in terms of cyclin E1 as a biomarkers of response with the ultimate aim of groping similar subsets in the future for therapeutic purposes. MethodsUsing automated tissue based assays, we assessed the expression and the amplification status of cyclin E1 in a cohort enriched for familial cancer mutations against each of the expression of cyclin E1 degradation associated proteins, i.e., FBXW7 and USP28, BRCA status and clinical outcome and in comparison with cyclin E1 high subsets of HGSOC. ResultsFor BLBC cyclin E1 overexpression but not CCNE1 amplification was found to be prognostic for overall survival. By comparison, both cyclin E1 expression and CCNE1 gene amplification are prognostic for HGSOC, which we establish with a meta-analysis of existing studies. We examined the nature of CCNE1 amplification to find that HGSOC has an enhanced degree of amplification of CCNE1 with 5.6x copies compared to 3.9x CCNE1 copies in BLBC, and this was associated with widely different expression of cyclin E1 protein. We find that BLBC cancers, with or without CCNE1 amplification, have similarities in dysregulation of regulators of cyclin E1 protein stability and those shared more features with the non-amplified cyclin E1 high HGSOC subset. Since cyclin E1 is only expressed at high levels in HGSOC and not BLBC, we examined the consequence of very high cyclin E1 expression in a BLBC cell line, MDA-MB-468. Moderate overexpression of cyclin E1 showed increased survival upon treatment with Paclitaxel, however very high expression of cyclin E1 gave less survival benefit. ConclusionsWe conclude that cyclin E1 high BLBC shared features with the non-amplified cyclin E1 high subset, high cyclin E1 expression associated with CCNE1 amplification gives rise to a unique phenotype in HGSOC, but otherwise the high expression of cyclin E1 protein may be a similar prognostic marker for BLBC and HGSOC. Legal entity responsible for the studyThe authors. FundingVentana Medical System.D. Aziz: Research grant / Funding (self), provided reagents and technical support: Ventana Medical System. P.M. Waring: Research grant / Funding (institution), Full / Part-time employment, provided reagents and technical support, worked for about 10 months for ventana during the project: Ventana Medical System. All other authors have declared no conflicts of interest.

Application of variant interpretation software to decipher pathogenicity of mutations for a molecular tumour board (MTB)

Abstract Background Comprehensive genomic profiling (CGP) of tumours using next generation sequencing (NGS) is increasingly used to guide management. Broad panel sequencing may increase treatment options by identifying more potential targets but generates large quantities of data. Therefore, determining actionable alterations is a challenge. To address this the MTB for the Tumour chARacterisation to Guide Experimental Targeted therapy (TARGET) trial at the Manchester Cancer Research Centre has evaluated a variant interpretation software package. Methods CGP of circulating tumour DNA (ctDNA) was undertaken using a somatic NGS 641 gene panel that included 24 cancer related genes and further exploratory genes. The variant allele frequency threshold was 2.5 and functional annotation of somatic variants used ANNOVAR. For cases with ≥2 mutations the vcf file was analysed using the Qiagen Clinical Insight platform (QCI) variant interpretation software. Results Over 18 months, 122 cases with a total of 1313 mutations were analysed by QCI. The variant interpretation provided information on the pathogenicity of the mutations and actionability, which was based on clinical trials identified by QCI. 11% of the total mutations were from the NGS 24 genes, of which 88% were pathogenic and of these 63% were actionable. The NGS exploratory genes accounted for the majority of mutations (89%). In the NGS exploratory genes a smaller percentage of mutations were pathogenic (17%) and of these 9% were actionable. The benefit for the MTB was the ability to focus on the relevant pathogenic mutations. Results were discussed in MTB meetings and pathogenic mutations captured in a digital tool, eTARGET. Conclusions Utilising variant interpretation software to identify pathogenic mutations from a broader CGP panel enabled more streamlined discussion and decision-making. Whilst a broader panel reveals numerically more mutations and the actionability rates are lower this is important to understand the biological relevance of single mutations and patterns of co-mutations. Applying a pathogenicity filter for ctDNA mutations using variant analysis software has scientific and potential clinical utility. Legal entity responsible for the study Cancer Research UK Manchester Institute, The University of Manchester. Funding Partly funded by AstraZeneca iDecide Programme (grant no 119106, C.D.). Disclosure S. Southam: Shareholder / Stockholder / Stock options: AstraZeneca; Shareholder / Stockholder / Stock options: GSK; Research grant / Funding (institution): AstraZeneca. All other authors have declared no conflicts of interest.

Plasma ctDNA RAS mutation analysis by digital polymerase chain reaction in patients with inoperable pancreatic cancer

Abstract Background Pancreatic ductal adenocarcinoma (PDAC) is associated with poor prognosis and high mortality worldwide. Despite the emergence of new therapeutic approaches for the treatment of inoperable PDAC, reliable prognostic/predictive biomarkers are lacking. PDAC harbours mutations in the KRAS gene in 80-90% of patients. The aim of the present study is the detection and quantification of RAS mutations in the ctDNA of patients with inoperable PDAC by BEAMing digital polymerase chain reaction (PCR), in order to (a) evaluate the correlation with tissue RAS mutation status in the tumour, (b) study the baseline prognostic significance and (c) study the predictive significance of ctDNA RAS mutation kinetics for the prediction of response to systemic/neoadjuvant therapy. Methods Plasma samples of patients with locally advanced or metastatic inoperable PDAC were collected at diagnosis and at 4-6 weeks after treatment initiation and were tested for 34 mutations in the KRAS and NRAS genes by BEAMing digital polymerase chain reaction (PCR). This method has a particularly high sensitivity and can accurately quantify mutated DNA as the Mutant Allele Fraction (MAF). To test the efficacy and validity of ctDNA for the detection of RAS mutations, RAS mutations will also be tested for on paired tumour tissue samples of the patients under study by PCR. Results So far, 38 plasma samples from 20 patients were available for analysis. At baseline, mutant ctDNA pertaining to exon 2 codon 12 of the KRAS gene was detected in all patients’ samples and, additionally, to exon 3 codon 61 of the NRAS gene in one of them. Comparison of plasma and tissue results revealed a high agreement of the RAS mutational profile in plasma ctDNA and tissue tumour DNA. Kinetics of the RAS mutational load during therapy and its prognostic or predictive value will be analysed with further cohort expansion. Conclusions This work highlights the potential role of liquid biopsies in the management of PDAC and, more specifically, will yield preliminary data on the role of RAS ctDNA profiling as a predictive and prognostic marker. Legal entity responsible for the study Department of Medical Oncology, University General Hospital of Ioannina. Funding Hellenic Society of Medical Oncology (HeSMO) and Society for Study of Clonal Heterogeneity of Neoplasia (EMEKEN). Disclosure G. Pentheroudakis: Research grant / Funding (institution), Support material was provided for this project: Sysmex Inostics. All other authors have declared no conflicts of interest.

502P - Epidermal growth factor receptor tyrosine kinase inhibitor treatment response in advanced non-small cell lung cancer with uncommon mutations: A multicenter observational study

BackgroundTreatment with epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) is the standard therapy for advanced non-small cell lung cancer (NSCLC) with common EGFR mutations. However, the efficacy of EGFR-TKIs in patients with uncommon EGFR mutations remains unclear. MethodsWe retrospectively surveyed a consecutive database of patients with NSCLC with EGFR mutations. We analyzed the data of patients with NSCLC with uncommon mutations, including single or compound mutations, who were treated with EGFR-TKIs between May 2016 and October 2018. ResultsData from 543 patients were collected from five institutions, among whom 23 had EGFR uncommon mutations. Twenty-one patients who were treated with any EGFR-TKIs were analyzed in this study, 18 of whom were treated with EGFR-TKIs as first-line therapy (gefitinib 5, erlotinib 3, afatinib 10 patients). In contrast, three patients underwent cytotoxic chemotherapy as first-line therapy and were treated with EGFR-TKIs as second- and third-line therapy (gefitinib 1, erlotinib 1, afatinib 1 patient). According to the Response Evaluation Criteria in Solid Tumors, the overall response rate was 56%, and the disease control rate was 78%. The median progression-free survival (PFS) was 14.0 months in all patients with uncommon mutations. The median PFS of patients who were treated with first and second generation EGFR-TKIs were 14.0 months (n = 10) and 7.3 months (n = 11), respectively. Moreover, the PFS of patients with the G719X mutation (n = 12, median PFS: 32.9 months) was longer than that of patients with the L861Q mutation (n = 4, median PFS: 11.1) and compound mutations (n = 4, median PFS 7.3 months). ConclusionsFirst and second generation EGFR-TKIs are effective treatments for patients with NSCLC with uncommon mutations. Notably, a greater favorable response was observed in patients with G719X mutations than those with L861Q and compound mutations. Clinical trial identificationUMIN000028989. Legal entity responsible for the studyFumio Imamura. FundingAstraZeneca. DisclosureF. Imamura: Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): AstraZeneca. All other authors have declared no conflicts of interest.

404O - A phase II single arm study of nivolumab and ipilimumab (Nivo/Ipi) in previously treated Classical Kaposi Sarcoma (CKS)

BackgroundCKS is an angioproliferative mesenchymal neoplasm causatively associated with human herpes virus 8 infection. Though recombinant IFNa is approved for treatment of AIDS-related KS, data is limited regarding the role of immune modulation. Based on favorable responses in viral-induced cancers, we hypothesized that CTLA-4 and PD-1 blockade can induce tumor regression in CKS. We present pre-planned interim analysis of a phase II study of Nivo/Ipi in previously treated (tx) progressive CKS. MethodsCKS pts with progressive disease after > 1 line of systemic tx and measurable disease by PET/CT and/or physical exam received nivolumab 240mg d1,15,28 and ipilimumab 1mg/kg d1 q42 days until progression or toxicity. The primary endpoint (EP) was overall response rate (ORR), secondary EP include 6-months progression free survival rate (PFS) and safety. Exploratory EP included PD-L1/MMR IHC, DNAseq (596 genes)/RNAseq (whole transcriptome) of tissue and matched blood to explore CKS genomic traits and IO correlates: TMB, MSI, dMMR, PD-L1, immune gene expression (PD1, CTLA4, etc) (Tempus Labs, Chicago, IL, USA). ResultsTwelve patients were enrolled and evaluable between Apr2018-Jun2019 and 11 were evaluable for response. Median age 72 (61-81). At a mFU of 6 months ORR was 45% (4 pts PR, 1 pt CR, 6 pts SD). mPFS was not reached, 6-mo PFS rate was 91% (1 out of 11 patients had PD). The safety profile was as expected with three patients with G2 toxicity (1 ALT/AST increase, 1 asymptomatic lipase increase) and two patients with G3 toxicity (1 colitis, 1 asymptomatic lipase increase). One SAE was reported (TIA considered not related to therapy) and treatment was discontinued in one pt (G2 LFT increase. maintaining CR 4 months after treatment discontinuation). Correlative results are available for four patients showing negative PDL1 IHC in all, low TMB, MS-S, but marked overexpression of CTLA4, PD1 and PDL1. ConclusionsThe interim analysis in this prospectively designed phase II study of nivo/ipi demonstrate promising activity in progressive CKS, with 45% ORR and a 6mo PFS rate of 91%. Toxicity profile as expected in this class of drugs. Correlative studies are preliminary but warrants further investigation into immune gene expression profiles. Clinical trial identificationNCT03219671. Legal entity responsible for the studyThe authors. FundingBMS Rabin Medical Center. DisclosureA. Zer: Honoraria (self), Research grant / Funding (institution): BMS; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Roche; Honoraria (self), Advisory / Consultancy: MSD; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: AZ. All other authors have declared no conflicts of interest.

64P - Entrectinib in locally advanced/metastatic ROS1 and NTRK fusion-positive non-small cell lung cancer (NSCLC): Updated integrated analysis of STARTRK-2, STARTRK-1 and ALKA-372-001

BackgroundEntrectinib is a systemic and central nervous system (CNS)-active potent inhibitor of ROS1 and TRKA/B/C. Primary data showed that entrectinib was tolerable and achieved high objective response rates (ORR) in patients (pts) with ROS1-positive (ROS1+), ROS1 inhibitor-naive NSCLC, and in pts with NTRK fusion-positive (NTRK+) NSCLC, including pts with baseline CNS disease. We present data from an additional 5 months of follow-up. MethodsPts with locally advanced/metastatic ROS1+ or NTRK+ tumors (with or without baseline CNS disease) confirmed by nucleic acid-based methods, enrolled in global phase 1/2 entrectinib trials (ALKA-372-001 [EudraCT 2012-000148-88], STARTRK-1 [NCT02097810], STARTRK-2 [NCT02568267]) were included. Disease burden was assessed per blinded independent central review (BICR) using RECIST v1.1, after cycle 1 (4 wks) then every 8 wks. Primary endpoints were ORR and duration of response (DOR) by BICR. Secondary endpoints included ORR and DOR in pts with or without baseline CNS disease, and safety. Intracranial (IC) ORR and DOR were evaluated in pts with baseline CNS disease. ResultsThere were 53 efficacy-evaluable pts with treatment-naïve, ROS1+ NSCLC and 10 pts with NTRK+ NSCLC. As of 30 Oct 2018 (additional 5 months’ follow-up), BICR ORR: ROS1+ 79.2% (95% CI 65.9–89.2) and NTRK+ 70.0% (95% CI 34.75–93.33) with complete responses in 5 (9.4%) pts and 1 (10.0%) pt, respectively. In ROS1+ NSCLC, median DOR: 24.6 mo (95% CI 12.6–34.8); in pts with and without baseline CNS disease, ORR was 73.9% (95% CI 51.6–89.8) and 83.3% (95% CI 65.3–94.4); IC ORR was 55.0% (95% CI 31.5–76.9); and median IC DOR was 12.9 mo (95% CI 5.6–not estimable). Additional efficacy for NTRK+ NSCLC pts will be presented. Entrectinib was well tolerated with a safety profile consistent with that previously reported; there were no new or unexpected safety findings. ConclusionsIn line with the primary data, in pts with ROS1+ and NTRK+ NSCLC after an additional 5 months of follow-up, entrectinib was well tolerated and showed clinically meaningful, durable systemic and intracranial responses. Clinical trial identificationALKA-372-001 [EudraCT 2012-000148-88] STARTRK-1 [NCT02097810] STARTRK-2 [NCT02568267]. Editorial acknowledgementMedical Writing support was provided by Laura Vergoz and Charlotte Kennerley, PhD of Gardiner-Caldwell Communications and was funded by F. Hoffmann-La Roche. Legal entity responsible for the studyF. Hoffman-La Roche. FundingF. Hoffman-La Roche. DisclosureF. de Braud: Advisory / Consultancy, Officer / Board of Directors: TizianaLife Sciences, BMS, Celgene, Novartis, Servier, Pharm Research Associated, Daiichi Sankyo, Ignyta, Amgen, Pfizer, Octimet Oncology, Incyte, Teofarma, Pierre Fabre, Roche, EMD Serono; Honoraria (self): BMS, Eli Lilly, Roche, Amgen, AstraZeneca, Gentili, Fondazione Menarini, Novartis, MSD, Ignyta, Bayer, Noema S.r.l., ACCMED, Dephaforum S.r.l., Nadirex, Roche, Biotechspert Ltd, PriME Oncology, Pfizer. S. Siena: Advisory / Consultancy: Amgen, Bayer, BMS, CheckmAb, Celgene, Daiichi-Sankyo, Incyte, Merck, Novartis, Roche-Genentech, and Seattle Genetics. F. Barlesi: Research grant / Funding (self), Travel / Accommodation / Expenses: Astra-Zeneca, Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, Eli Lilly Oncology, F. Hoffmann–La Roche Ltd, Novartis, Merck, MSD, Pierre Fabre, Pfizer and Takeda; Research grant / Funding (institution): AbbVie, ACEA, Amgen, Astra-Zeneca, Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, Eisai, Eli Lilly Oncology, F. Hoffmann–La Roche Ltd, Genentech, Ipsen, Ignyta, Innate Pharma, Loxo, Novartis, Medimmune, Merck, MSD, Pierre Fabre, Pfizer, Sanofi-Aventis; Non-remunerated activity/ies: Principal Investigator for Astra-Zeneca, BMS, Merck, Pierre Fabre and Roche sponsored trials (or ISR). A. Drilon: Honoraria (self), Advisory / Consultancy: Ignyta/Genentech/Roche Loxo/Bayer/Lilly Takeda/Ariad/Millenium TP Therapeutics AstraZeneca Pfizer Blueprint Medicines Helsinn Beigene BergenBio Hengrui Therapeutics Exelixis Tyra Biosciences Verastem MORE Health; Research grant / Funding (institution): Pfizer Exelixis GlaxoSmithKlein Teva Taiho PharmaMar; Research grant / Funding (self): Foundation Medicine; Licensing / Royalties: Wolters Kluwer; Travel / Accommodation / Expenses: Merck - Food/Beverage Puma - Food/Beverage; Honoraria (self): Medscape, OncLive, PeerVoice, Physicians Education Resources, Targeted Oncology, Research to Practice. B. Simmons: Full / Part-time employment: Roche (Genentech). X. Huang: Shareholder / Stockholder / Stock options, Full / Part-time employment: Genentech. S. Osborne: Full / Part-time employment: Roche. R.C. Doebele: Shareholder / Stockholder / Stock options: Rain Therapeutics; Advisory / Consultancy: Chair of Scientific Advisory Board for Rain Therapeutics; Honoraria (self): Guardant; Advisory / Consultancy: Pfizer, Trovagene, Ariad, Takeda, AstraZeneca, Genentech/Roche, Ignyta, Loxo, Rain.; Research grant / Funding (self): Ignyta, Loxo, Mirati; Licensing / Royalties: Abbott Molecular, Rain Therapeutics, GVKbio, Chugai, Loxo, Ignyta, Genentech, Ariad, Foundation Medicine, Black Diamond.

507P - Immune checkpoint inhibitors for patients acquired resistance to tyrosine kinase inhibitors with EGFR mutated non-small cell lung cancer: A multicenter retrospective study

BackgroundEpidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) represent a highly effective treatment for advanced non-small-cell lung cancer (NSCLC) with active mutations of the EGFR gene. However, most patients develop acquired resistance to EGFR-TKIs. The T790M mutation has been found to be associated with almost 50% of patients who acquired resistance to first- or second- generation EGFR-TKIs. The role of immune checkpoint inhibitors (ICIs) for patients without a T790M mutation or patients with a T790M mutation who have progressed after T790M inhibitor therapy remains unclear. MethodsWe retrospectively evaluated the clinical effects of ICIs for EGFR mutated non- squamous NSCLC patients who were treated after developing resistance to first- or second- generation EGFR-TKIs at five institutions in Japan. Patients treated with EGFR-TKI between May 2016 and October 2018 were enrolled. Patients who used a third-generation EGFR-TKI before using prior generations were excluded. ResultsOf the 58 patients identified, 21 were positive for T790M, and all were developing resistance to both third- and prior generation EGFR-TKIs. The objective response and disease control rates for ICIs were 13.8% and 55.2% respectively, and T790M-negative patients had numerically greater responses than T790M-positive patients (16.2% versus 9.5% and 62.1% versus 42.9%, p = 0.70 and p = 0.18, respectively). T790M-negative patients were associated with a clinically meaningfully longer median progression-free survival than were T790M-positive patients (4.4 months versus 1.8 months, p = 0.061). ConclusionsT790M-negative status after development of acquired resistance to EGFR-TKIs tended to be associated with benefit from immunotherapy in this retrospective analysis. Clinical trial identificationUMIN000028989. Legal entity responsible for the studyThe authors. FundingAstraZeneca K. K. DisclosureF. Imamura: Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): AstraZeneca. All other authors have declared no conflicts of interest.

443P - The effectiveness of individual nutritional counselling for patients with advanced cancer undergoing chemotherapy: A preliminary study

BackgroundMany patients with advanced cancer have difficulty eating and need nutritional support. This preliminary study examined the effects of individual nutritional counselling sessions on quality of life (QOL), nutritional status and eating-related distress (ERD) among patients with advanced cancer undergoing chemotherapy. MethodsA survey was conducted in the outpatient chemotherapy center at a university hospital in Tokyo, Japan. Patients were offered individual nutritional counselling by a dietitian specializing in cancer. For the participants who completed at least two counselling sessions, we evaluated changes in their QOL (EORTC-QLQ C30), symptoms and nutritional status (PG-SGA SF) and ERD. Data were analyzed using the Wilcoxon signed rank test to compare mean differences. The ethics committee approved the study (Ethical Review No. M2015-578). ResultsOf 42 patients who applied for nutritional counselling, 8 underwent two sessions. The mean age was 70.6±5.9 years and females accounted for 25.0%. Primary diseases were gastrointestinal cancer (n = 6) and lung cancer (n = 2). “Although family members and friends recommend foods to me, I am just confused” on the ERD and “fatigue” on the EORTC-QLQ C30 significantly improved (p < 0.05, respectively). There were no significant differences in the other items of the EORTC-QLQ C30 or symptoms and nutritional status on the PG-SGA SF. In total, 34 patients did not receive nutritional counselling twice because of death, changing hospitals, disease progression, or lack of coordination among researchers. ConclusionsOur study suggests that individual nutritional counselling is slightly beneficial for patients with advanced cancer receiving chemotherapy. Further research is needed to demonstrate the effectiveness of this intervention in the future. Clinical trial identificationUMIN registration No. UMIN000021540; was carried out prior to the start of the study. Legal entity responsible for the studyTokyo Medical and Dental University. FundingSasakawa Memorial Health Foundation. DisclosureS. Koshimoto: Research grant / Funding (institution): Sasakawa Memorial Health Foundation. All other authors have declared no conflicts of interest.

531TiP - CANOPY-A: A phase III, placebo-controlled study of canakinumab as adjuvant therapy in patients (pts) with surgically resected NSCLC

531TiP - CANOPY-A: A phase III, placebo-controlled study of canakinumab as adjuvant therapy in patients (pts) with surgically resected NSCLC

481P - Updated survival outcomes of the phase II study of low starting dose of afatinib as first-line treatment in patients with EGFR mutation-positive non-small cell lung cancer (KTORG1402)

BackgroundThe phase II trial evaluated the efficacy and safety of low starting dose of afatinib in patients with advanced epidermal growth factor receptor (EGFR) mutation-positive non-small cell lung cancer (NSCLC). In primary analysis, progression free survival (PFS) met the primary endpoint. Here, we report updated survival outcomes. MethodsThis study was a multi-center, single-arm, open-label phase II trial. Treatment-naïve patients with common EGFR mutation-positive NSCLC were treated with afatinib starting at a dose of 20 mg/day. If tolerated, the dose is increased by 10- mg increments up to 50mg/day. PFS and overall survival (OS) were re-evaluated at the final data cut-off point (October 2018). Results46 patients were enrolled. Median age was 73 years (range, 43-86). The median follow-up was 31.9 months. Median PFS of entire population was 15.2 months (95% confident interval (CI), 13.2–21.2), and 2-year PFS rate was 27.7% (95% CI, 17.1-44.7). Median OS was not estimated, and the 2-year overall survival was 76.1% (95% CI, 64.7-89.5). Performance status (PS) and existence of brain metastases at study entry were revealed to have significant impact on the survival with Cox-regression test. Among patients with disease progression on afatinib (n = 36), rebiopsy was performed on 29 patients, and T790M was proved positive on 14 patients. Twelve out of the 14 patients positive for T790M received osimertinib as a second line therapy. Median time from enrollment to progression on the osimertinib (PFS2) was 32.6 months (95%CI: 20.5-NE). ConclusionsLow starting dose afatinib therapy demonstrated promising OS outcome. PS and existence of brain metastases were predictive factors of this therapy. Clinical trial identificationUMIN 000016444. Legal entity responsible for the studyKyoto Thoracic Oncology Research Group. FundingHas not received any funding. DisclosureT. Yokoyama: Honoraria (self): Boehringer Ingelheim Japan; Honoraria (self): Chugai Pharmaceutical; Honoraria (self): Taiho Pharmaceutical; Honoraria (self): Bristol-Myers Squibb; Honoraria (self): Novartis; Honoraria (self): AstraZeneca; Honoraria (self): Ono pharmaceutical; Honoraria (self): MSD. H. Yoshioka: Honoraria (self): Boehringer Ingelheim; Honoraria (self): Chugai Pharmaceutical; Honoraria (self): Taiho Pharmaceutical; Honoraria (self): Eli Lilly Japan; Honoraria (self): Bristol-Myers Squibb; Honoraria (self): Novartis; Honoraria (self): Merck Serono; Honoraria (self): Kyowa Kirin; Honoraria (self): AstraZeneca; Honoraria (self): Ono pharmaceutical; Honoraria (self): Daiichi Sankyo; Honoraria (self): MSD. D. Fujimoto: Honoraria (self), Research grant / Funding (self): AstraZeneca KK; Honoraria (self): Ono Pharmaceutical Co Lt; Honoraria (self): Bristol-Myers Squibb Co Ltd; Honoraria (self): Taiho Pharmaceutical Co ; Honoraria (self), Research grant / Funding (self): Chugai Pharmaceutical Co Ltd; Honoraria (self): MSD KK; Honoraria (self): Boehringer Ingelheim Japan Inc; Honoraria (self): Eli Lilly Japan KK. K. Hirano: Honoraria (self): Boehringer Ingelheim Japan. T. Ishida: Honoraria (self): Boehringer Ingelheim Japan. T. Hirai: Research grant / Funding (institution): Boehringer Ingelheim Japan. All other authors have declared no conflicts of interest.

422O - Needs of cancer patients in an Asian setting

BackgroundThe ASEAN Costs in Oncology (ACTION) Study had previously reported high levels of psychological distress among cancer survivors in the region. We aimed to gain an in-depth understanding of the (unmet) needs in Asian patients living with cancer in a middle-income setting. MethodsTwenty focus group discussions were conducted among patients with breast, cervical, prostate or colorectal cancer (N = 102) at five tertiary Malaysian hospitals (public, academic, private). Thematic analysis was performed. ResultsSeven themes were identified. (1) Information; participants emphasized the need for physicians to provide detailed information on latest treatment options. Besides medical advice, informational needs on diet, and traditional and complementary medicine remained unmet.”(2) Psychosocial; “coping with cancer diagnosis” and “fear of recurrence” were major issues. The need for “support groups” as a source of informational and psychological support was raised. It was also stressed that increased “public awareness” was needed to circumvent the stigma against cancer survivors. (3) Physical; patients stressed on concerns regarding “side effects of treatment”, including “fertility”, “sexuality” and “sexual function”. (4) Practical support; major unmet needs in “transportation and parking” and “comfortable facilities” in hospitals were reported. (5) Financial; The burden of “out-of-pocket” expenses, “paying first then claim” and the costs of “essential items” e.g. breast prosthesis, colostomy bags were highlighted. (6) Systemic; Issues with “accessibility and eligibility of financial aid” were frustrating to needy patients. The need for an “integrated care system” where different departments work together to avoid repeated tests, and consolidate hospital appointments into the same day was repeatedly raised. (7) Employment; Patients needed “job security” and “cancer leave”. Lack of work-related flexibility, and discrimination at workplace were voiced out. ConclusionsThere seems to be an overarching theme of limited patient-centered healthcare during the survivorship period in Asian settings. These findings underscore the need for holistic cancer survivorship services that address wider aspects of wellbeing including the urgent need for patient navigation programs. Legal entity responsible for the studyThe authors. FundingPharmaceutical Association of Malaysia. DisclosureN. Bhoo-Pathy: Research grant / Funding (institution): Pharmaceutical Association of Malaysia. All other authors have declared no conflicts of interest.

Customisation of therapeutic strategy in metastatic colorectal cancer by use of liquid biopsies: Updated results of an observational study

Abstract Background Metastatic colorectal cancer (mCRC) affects 20% of patients diagnosed with colorectal malignancy and is a major cause of cancer-related death worldwide. Epidermal growth factor receptor (EGFR) blockade with monoclonal antibodies, in combination with chemotherapy, has been shown to benefit patients with RAS wild type tumours. Our project explores real-time assessment of mCRC mutational profile, by use of two methods of circulating tumour DNA (ctDNA) analysis, and its potential to better guide therapeutic decisions. Methods Baseline plasma samples were prospectively collected from 64 patients at diagnosis of mCRC and at disease progression. Analysis of KRAS and NRAS exons 2, 3 and 4 was performed with the highly sensitive BEAMing digital polymerase chain reaction (PCR) and next-generation sequencing (NGS). In a subgroup of patients, with baseline RAS mutations, mid-treatment liquid biopsies were performed, to assess the change in ctDNA RAS mutational load as response biomarker. Mutational status of matched tumour tissue, generated with a conventional PCR-based method or NGS, was prospectively recorded for all patients. Results Preliminary results of our study, presented at the EACR-ESMO Joint Conference on Liquid Biopsies, have shown detection of RAS mutations in the ctDNA from over 50% of patients, with a satisfactory plasma-tissue concordance. An update on these data, as well as on the predictive value of ctDNA RAS mutational load dynamics, will be presented. Disease progression data, in light of baseline RAS status, will also be revisited, as almost 20% of patients have progressed on first-line treatment. In addition, the concordance of BEAMing and NGS as liquid biopsy methods is to be studied in a subgroup of patients with available plasma samples. Conclusions The larger sample size and the maturity of this updated analysis further add on the value of our work, the results of which remain in line with the literature. Our study serves as a real-life example of the potential of liquid biopsies to detect RAS mutation status changes, allowing for precision-oncology therapeutic approaches in the continuum of care of metastatic colorectal cancer patients. Legal entity responsible for the study Department of Medical Oncology, University General Hospital of Ioannina. Funding Amgen, Hellenic Society of Medical Oncology (HeSMO) and Society for Study of Clonal Heterogeneity of Neoplasia (EMEKEN). Disclosure G. Pentheroudakis: Research grant / Funding (institution): Amgen. All other authors have declared no conflicts of interest.