A phase II study of bemcentinib (BGB324), a first-in-class highly … · 2019. 6. 2. · pharmDx...

1
Poster # 421 Promising clinical activity, particularly in patients with AXL positive disease, independent of PD-L1 expression Background & objective This poster was presented at the ASCO Annual Meeting 2019 Chicago, IL (May 31 - 04 Jun) This presentation is the intellectual property of the presenter. Contact them at [email protected] for permission to reprint and/or distribute. Mutations* n % None 35 76% KRAS 6 13% TP53 2 4% ERBB2 1 2% EGFR 1 2% Other/Unknown 2 4% Best response to most recent treatment n % CR 2 4% PR 17 37% SD 10 22% PD 12 26% Unknown 5 11% Disease Characteristics * May be overlap between individual patients The studied population was predominently PD-L1 negative (53%) patients who are less likely to benefit from pembrolizumab monotherapy treatment. The studied population was predominently AXL positive (58%) patients. Promising clinical activity continues to be seen overall, particularly in patients with AXL positive tumours including those with low or no PD-L1 expression. Conclusions The median overall survival has surpassed what has been shown historically in 2L treatment with PD-1 inhibitor monotherapy The combination treatment of bemcentinib and pembrolizumab was well-tolerated. 1 Herbst, et al. The Lancet (2016). Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial. 2 Garon, et al. NEJM (2015). Pembrolizumab for the Treatment of Non–Small-Cell Lung Cancer. 3 Davidsen, et al. Springer Publishing (2017). The Role of Axl Receptor Tyrosine Kinase in Tumor Cell Plasticity and Therapy Resistance. 4 Shieh, et al. Neoplasia (2005). Expression of axl in lung adenocarcinoma and correlation with tumor progression. 5 Leighl, et al. Lancet (2019). Pembrolizumab in patients with advanced non-small-cell lung cancer (KEYNOTE-001): 3-year results from an open-label, phase 1 study. References Safety * Preferred terms include: Alanine aminotransferase increased, Aspartate aminotransferase increased and Transaminases increased. All events were reversible No grade 5 TRAEs were reported. Most frequent TRAEs (occurring in >10% of dosed patients) n = 46 Preferred term All grades Grades 3 n % n % Transaminase increase* 16 35% 6 13% Asthenia / Fatigue 14 30% 2 4% Diarrhoea 12 26% 0 0% Nausea 6 13% 0 0% Anaemia 5 11% 1 2% Decreased appetite 5 11% 0 0% 1 Vall d'Hebron University Hospital, Barcelona, Spain; 2 Oslo University Hospital, Oslo, Norway; 3 Hospital Clinic, Barcelona, Spain; 4 Hospital 12 de Octubre, Madrid, Spain; 5 Hospital Germans Trias i Pujol, Badalona, Spain; 6 Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain; 7 Hospital Virgen de la Victoria, Málaga, Spain; 8 Hospital del Mar, Barcelona, Spain; 9 University Hospital A Coruña , A Coruña, Spain; 10 Kings College London, Guy’s Hospital, London, UK; 11 Medical College of Wisconsin Affiliated Hospitals, Menomonee Falls, WI; 12 Complejo Hospitalario de Jaén, Jaén, Spain; 13 BerGenBio ASA, Bergen, Norway; 14 Merck & Co., Inc., Kenilworth, NJ; 15 The Christie NHS Foundation Trust and The University of Manchester, Manchester, UK A phase II study of bemcentinib (BGB324), a first-in-class highly selective AXL inhibitor, with pembrolizumab in patients with advanced NSCLC: OS for stage I and preliminary stage II efficacy. Enriqueta Felip 1 , Paal Brunsvig 2 , Nuria Vinolas 3 , Santiago Ponce Aix 4 , Enric Carcereny Costa 5 , Manuel Dómine Gomez 6 , Jose Manuel Trigo Perez 7 , Edurne Arriola 8 , Rosario Garcia Campelo 9, James F. Spicer 10 , Jonathan Robert Thompson 11 , Ana Laura Ortega Granados 12 , Robert Holt 13 , Katherine Lorens 13 , James B. Lorens 13 , Muhammad Shoaib 13 , Abdul Siddiqui 13 , Emmett V. Schmidt 14 , Michael Jon Chisamore 14 , Matthew Krebs 15 P P bemcentinib immune evasion drug resistance Gas6 AXL Bemcentinib isa highly selective, potent, orally available, AXL kinase inhibitor Selectivity Profile phase I phase II phase III Bemcentinib – selective AXL kinase inhibitor NSCLC adenocarcinoma mutation driven adenocarcinoma Melanoma AML/MDS Combination with low dose chemo + Single agent (NCT02488408) Combination with erlotinib (NCT02424617) Combination with pembrolizumab (NCT03184571) Combination with docetaxel (NCT02922777) Combination with SoC therapies, incl. CPIs (NCT02872259) Bemcentinib is being developed as monotherapy and in combination with immune, targeted and chemotherapy in NSCLC, AML/MDS and melanoma. Kinase Kinome Scan (Kd) nM Fold Axl 0.4 1 Mer 100 250 Tyro >1000 >1000 80% 60% -60% -40% -20% 0% 20% 40% -80% - - - - - - - ++ + + + + NA + + NA NA NA - - + + + NA ++ + - - - - NA PD PD PD PD PD PD PR PR PR PR PR PR PR Antitumour activity: Change in tumour size from baseline (by AXL IHC) AXL Positive AXL Negative Not evaluable PR PR PR SD SD SD PD PD PD PD PD PD PD SD SD SD SD SD SD SD SD SD best % change in sum target lesions PD-L1 Status + - - + (-): negative; (+): positive (TPS = 1-49%; (++): strong positive (TPS ≥50%) (NA): PD-L1 status not available Screened: n = 74 Received at least 1 dose of combination: n=46 Biomarker evaluable AXL Status: n=33 Discontinued n=30 Ongoing n=16 Patient disposition, stages I + II PD-L1 Status: n=38 29 with at least 1 radiological assessment At least 1 radiological assessment n=35 Progression without radiological assessment n=3 No response assessment n=8 28 with at least 1 radiological assessment 15 PD 7 AE 3 death 3 investigator decision 2 withdrawal of consent AXL Negative AXL Positive Positive (TPS 1-49%) Negative (TPS ≤1%) 58% 42% Strong Positive (TPS ≥50%) 53% 39% 8% 58% 42% 0% 20% 30% 40% 50% 60% -20% -40% -60% -80% 15.0 35.0 45.0 55.0 65.0 25.0 AXL Negative AXL Positive AXL status not available or not yet received Ongoing patient Of the 46 patients who have been dosed, 35 had a radiological assessment (shown above). Of the 11 patients not shown, 3 had PD without evaluable radiological assessment, and 8 patients either had no response assessment or were early in treatment. % change Time since first dose (weeks) Change in sum of target lesions over time, by patient 5.0 Methods Immunohistochemistry: AXL IHC was performed by Indivumed on pre-treatment FFPE samples using a BerGenBio proprietary immunohistochemistry assay PD-L1 status was determined using a 1% cutoff by IHC using the PD-L1 IHC 22C3 pharmDx assay (Agilent, Carpinteria, CA, USA). Scoring was recorded as percentage of PD-L1-positive tumor cells over total tumor cells in the denominator (TPS). Tumour assessments: Target lesions were assessed as per RECIST v1.1. Assessments were carried out every 9 weeks. Adverse events: Adverse events were assessed by CTCAE v4.03 1.2 1 0.8 0.6 0.4 0.2 0 2 4 6 8 0 14 12 10 16 18 Time (Months) fraction at risk Median (CI 95%): 12.2 months (6.2-NR) 12-mo OS: 54.2% reference value 5 : KN-001, all previously -treated pts reference value 5 : KN-001, previously-treated pts, non-squamous reference value 5 : KN-001, previously-treated pts, TPS 1-49% reference value 5 : KN-001, previously-treated pts, TPS <1% Median overall survival in stage I patients (n=24) OS (overall, n=24) Median OS Bemcentinib + Pembrolizumab Pembrolizumab Patient demographics Median Age (range) 64.5 (39-82) ECOG at screen 0 (%) 22 (48%) 1 (%) 24 (52%) ≥2 (%) 0 Sex Female (%) 18 (39%) Male (%) 28 (61%) Ethnicity Hispanic or Latino (%) 9 (20%) Not Hispanic or Latino (%) 37 (80%) Race White (%) 43 (94%) Asian (%) 2 (4%) Other (%) 1 (2%) Smoking status Smoker (%) 8 (17%) Ex-smoker (%) 27 (59%) Never smoked (%) 10 (22%) Unknown (%) 1 (2%) Pack years Median 36.5 Range 0.5-100 BerGenBio ASA Jonas Lies vei 91 5009 Bergen NORWAY BerGenBio Ltd 1 Robert Robinson Ave OX4 4GA Oxford, UK www.bergenbio.com @BGenBio [email protected] +47 559 61 159 Contact 2nd line advanced adeno NSCLC • IO naive • prior platinum therapy • measurable disease • fresh tissue biopsy • PD-L1 +ve and -ve • AXL +ve and -ve Single arm bemcentinib, 200mg daily + pembrolizumab, 200mg q3wks Interim analysis ( n = 24 patients) Final analysis (n = 48 patients) • Histopathologically or cytologically documented Stage IV adenocarcinoma of the lung • Has disease progression on or after a prior platinum-containing chemotherapy • Measurable disease as defined by RECIST 1.1 • Provision of suitable fresh tumour tissue for the analysis of AXL kinase expression and PD-L1 expression • Eastern Cooperative Oncology Group (ECOG) performance score 0 or 1 • Not received more than one prior line of chemotherapy for advanced or metastatic adenocarcinoma of the lung • No prior therapy with an immunomodulatory agent • No known active central nervous system (CNS) metastases and/or carcinomatous meningitis • No recent or ongoing systemic steroid therapy NCT03184571: Phase II clinical trial of selective AXL inhibitor bemcentinib in combination with pembrolizumab Simon-like two stage design enrolling up to 48 patients Key inclusion and exclusion criteria Assessments - efficacy & safety • Response was assessed every 9 weeks per RECIST v1.1 • Adverse events were assessed by CTCAE v4.03 • Evaluable: ≥1 dose of study treatment as of data cutoff (23 Apr 2019) Biomarker analysis • PD-L1 and AXL expression by IHC • Soluble protein biomarkers by liquid biopsy • Immune cell populations by multi-spectral imaging Endpoints Primary: ORR Secondary: DCR, DoR, TtP, Survival at 12 months, response by biomarker expression Study rationale Anti PD-1 therapies in second line metastatic non-small cell lung cancer (NSCLC) AXL receptor tyrosine kinase and selective AXL inhibitor bemcentinib • Pembrolizumab as a single agent is indicated for the treatment of patients with metastatic NSCLC with disease progression on or after platinum-containing chemotherapy 1,2 • KEYNOTE-001 5 , a multi-cohort phase I study designed to define and validate expression levels of PD-L1 associated with the likelihood of clinical benefit, showed that pembrolizumab monotherapy efficacy is correlated with PD-L1 levels • Novel combination treatment strategies are needed to improve efficacy of pembrolizumab while limiting additive toxicity • AXL is a receptor tyrosine kinase expressed on tumour and immune cells and a member of the TAM family (Tyro-AXL-Mer) of kinases • AXL is overexpressed in response to a hostile tumour micro-environment and drives a tumour survival programme including immune escape, anti-tumour therapy resistance & metastasis 3 • AXL is a negative prognostic factor in a multitude of cancers including NSCLC 4 • Bemcentinib, a first-in-class highly selective, potent, and orally bioavailable inhibitor of AXL, has been shown to potently improve the efficacy of checkpoint blockade in murine pre-clinical models of NSCLC n PR SD PD ORR (%) CBR (%) Overall** 35 10 12 13 29% 63% AXL 28 Positive* 15 6 4 5 40% 67% Negative 13 2 5 6 15% 54% PD-L1 29 PD-L1 strong positive (TPS 50%) 2 1 0 1 50% 50% PD-L1 positive (TPS 1-49%) 12 3 4 5 25% 58% PD-L1 negative (TPS <1%) 15 4 5 6 27% 60% *Any AXL expression as measured by IHC (cut off in development) **All patients with radiological assessments included (n=35)

Transcript of A phase II study of bemcentinib (BGB324), a first-in-class highly … · 2019. 6. 2. · pharmDx...

Page 1: A phase II study of bemcentinib (BGB324), a first-in-class highly … · 2019. 6. 2. · pharmDx assay (Agilent, Carpinteria, CA, USA). Scoring was recorded as percentage of PD-L1-positive

Poster # 421

Promising clinical activity, particularly in patients with AXL positive disease, independent of PD-L1 expressionBackground & objective

This poster was presented at the ASCO Annual Meeting 2019 Chicago, IL (May 31 - 04 Jun)This presentation is the intellectual property of the presenter. Contact them at [email protected] for permission to reprint and/or distribute.

Mutations* n %None 35 76%KRAS 6 13%TP53 2 4%ERBB2 1 2%EGFR 1 2%Other/Unknown 2 4%Best response to most recent treatment

n %

CR 2 4%PR 17 37%SD 10 22%PD 12 26%Unknown 5 11%

Disease Characteristics

* May be overlap between individual patients

The studied population was predominently PD-L1 negative (53%) patients who are less likely to benefit from pembrolizumab monotherapy treatment.The studied population was predominently AXL positive (58%) patients.

Promising clinical activity continues to be seen overall, particularly in patients with AXL positive tumours including those with low or no PD-L1 expression.

ConclusionsThe median overall survival has surpassed what has been shown historically in 2L treatment with PD-1 inhibitor monotherapy

The combination treatment of bemcentinib and pembrolizumab was well-tolerated.

1Herbst, et al. The Lancet (2016). Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial.2Garon, et al. NEJM (2015). Pembrolizumab for the Treatment of Non–Small-Cell Lung Cancer.3Davidsen, et al. Springer Publishing (2017). The Role of Axl Receptor Tyrosine Kinase in Tumor Cell Plasticity and Therapy Resistance.4Shieh, et al. Neoplasia (2005). Expression of axl in lung adenocarcinoma and correlation with tumor progression.5Leighl, et al. Lancet (2019). Pembrolizumab in patients with advanced non-small-cell lung cancer (KEYNOTE-001): 3-year results from an open-label, phase 1 study.

References

Safety

* Preferred terms include: Alanine aminotransferase increased, Aspartate aminotransferase increased and Transaminases increased. All events were reversibleNo grade 5 TRAEs were reported.

Most frequent TRAEs (occurring in >10% of dosed patients)n = 46

Preferred term All grades Grades 3

n % n %

Transaminase increase* 16 35% 6 13%

Asthenia / Fatigue 14 30% 2 4%

Diarrhoea 12 26% 0 0%

Nausea 6 13% 0 0%

Anaemia 5 11% 1 2%

Decreased appetite 5 11% 0 0%

1 Vall d'Hebron University Hospital, Barcelona, Spain; 2 Oslo University Hospital, Oslo, Norway; 3 Hospital Clinic, Barcelona, Spain; 4 Hospital 12 de Octubre, Madrid, Spain; 5 Hospital Germans Trias i Pujol, Badalona, Spain; 6 Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain; 7 Hospital Virgen de la Victoria, Málaga, Spain; 8 Hospital del Mar, Barcelona, Spain; 9 University Hospital A Coruña , A Coruña, Spain; 10 Kings College London, Guy’s Hospital, London, UK; 11Medical College of Wisconsin Affiliated Hospitals, Menomonee Falls, WI; 12 Complejo Hospitalario de Jaén, Jaén, Spain; 13 BerGenBio ASA, Bergen, Norway; 14 Merck & Co., Inc., Kenilworth, NJ; 15 The Christie NHS Foundation Trust and The University of Manchester, Manchester, UK

A phase II study of bemcentinib (BGB324), a first-in-class highly selective AXL inhibitor, with pembrolizumab in patients with advanced NSCLC: OS for stage I and preliminary stage II efficacy.Enriqueta Felip1, Paal Brunsvig2, Nuria Vinolas3, Santiago Ponce Aix4, Enric Carcereny Costa5, Manuel Dómine Gomez6, Jose Manuel Trigo Perez7, Edurne Arriola8, Rosario Garcia Campelo9, James F. Spicer10, Jonathan Robert Thompson11, Ana Laura Ortega Granados12, Robert Holt13, Katherine Lorens13, James B. Lorens13, Muhammad Shoaib13, Abdul Siddiqui13, Emmett V. Schmidt14, Michael Jon Chisamore14, Matthew Krebs15

P P

bemcentinib immune evasiondrug resistance

Gas6

AXL

Bemcentinib isa highly selective, potent, orally available, AXL kinase inhibitor

Selectivity Profilephase I phase II phase IIIBemcentinib – selective AXL kinase inhibitor

NSCLC

adenocarcinoma

mutation driven

adenocarcinoma

Melanoma

AML/MDS Combination with low dose chemo + Single agent (NCT02488408)

Combination with erlotinib (NCT02424617)

Combination with pembrolizumab (NCT03184571)

Combination with docetaxel (NCT02922777)

Combination with SoC therapies, incl. CPIs (NCT02872259)

Bemcentinib is being developed as monotherapy and in combination with immune, targeted and chemotherapy in NSCLC, AML/MDS and melanoma.

KinaseKinome Scan (Kd)

nM Fold

Axl 0.4 1

Mer 100 250

Tyro >1000 >1000

80%

60%

-60%

-40%

-20%

0%

20%

40%

-80%

- + - - - -- -+++++ + NA ++ NA NA NA-- +++ NA ++ + - - -- NA

PDPD

PDPDPDPD

PRPR

PRPRPRPRPR

Antitumour activity: Change in tumour size from baseline (by AXL IHC)

AXL Positive

AXL Negative

Not evaluable

PRPR

PR

SDSDSD

PDPD

PDPD

PDPD

PD

SD

SDSDSDSDSD

SD

SD SD

best

% c

hang

e in

sum

targ

et le

sion

s

PD-L1 Status

+--+

(-): negative; (+): positive (TPS = 1-49%; (++): strong positive (TPS ≥50%) (NA): PD-L1 status not available

Screened: n = 74

Received at least 1 dose of combination:

n=46

Biomarker evaluable

AXL Status: n=33

Discontinuedn=30

Ongoingn=16

Patient disposition, stages I + II

PD-L1 Status: n=3829 with at least 1 radiological

assessment

At least 1 radiological assessment

n=35

Progression without radiological assessment

n=3

No response assessmentn=8

28 with at least 1 radiological assessment

15 PD7 AE

3 death3 investigator decision

2 withdrawal of consent

AXL Negative

AXL Positive

Positive(TPS 1-49%)

Negative(TPS ≤1%)

58% 42%

Strong Positive(TPS ≥50%)

53%39%

8%

58%42%

0%

20%

30%

40%

50%

60%

-20%

-40%

-60%

-80%

15.0 35.0 45.0 55.0 65.025.0

AXL Negative

AXL Positive

AXL status not available or not yet received

Ongoing patient

Of the 46 patients who have been dosed, 35 had a radiological assessment (shown above). Of the 11 patients not shown, 3 had PD without evaluable radiological assessment, and 8 patients either had no response assessment or were early in treatment.

% c

hang

e

Time since first dose (weeks)

Change in sum of target lesions over time, by patient

5.0

Methods

Immunohistochemistry: AXL IHC was performed by Indivumed on pre-treatment FFPE samples using a BerGenBio proprietary immunohistochemistry assay PD-L1 status was determined using a 1% cutoff by IHC using the PD-L1 IHC 22C3 pharmDx assay (Agilent, Carpinteria, CA, USA). Scoring was recorded as percentage of PD-L1-positive tumor cells over total tumor cells in the denominator (TPS).

Tumour assessments: Target lesions were assessed as per RECIST v1.1. Assessments were carried out every 9 weeks.

Adverse events: Adverse events were assessed by CTCAE v4.03

1.2

1

0.8

0.6

0.4

0.2

02 4 6 80 141210 16 18

Time (Months)

fract

ion

at ri

sk

Median (CI 95%): 12.2 months (6.2-NR)12-mo OS: 54.2%

reference value5: KN-001,all previously -treated ptsreference value5: KN-001, previously-treated pts, non-squamousreference value5: KN-001, previously-treated pts, TPS 1-49%reference value5: KN-001, previously-treated pts, TPS <1%

Median overall survival in stage I patients (n=24)

OS (overall, n=24)

Median OS

Bemcentinib +

PembrolizumabPembrolizumab

Patient demographics

Median Age (range) 64.5 (39-82)ECOG at screen

0 (%) 22 (48%)1 (%) 24 (52%)

≥2 (%) 0Sex

Female (%) 18 (39%)Male (%) 28 (61%)

EthnicityHispanic or Latino (%) 9 (20%)

Not Hispanic or Latino (%) 37 (80%)Race

White (%) 43 (94%)Asian (%) 2 (4%)Other (%) 1 (2%)

Smoking statusSmoker (%) 8 (17%)

Ex-smoker (%) 27 (59%)Never smoked (%) 10 (22%)

Unknown (%) 1 (2%)Pack years

Median 36.5Range 0.5-100

BerGenBio ASAJonas Lies vei 915009 BergenNORWAY

BerGenBio Ltd1 Robert Robinson AveOX4 4GA Oxford, UK

www.bergenbio.com @BGenBio

[email protected]+47 559 61 159

Contact

2nd line advanced adeno NSCLC• IO naive • prior platinum therapy• measurable disease• fresh tissue biopsy• PD-L1 +ve and -ve• AXL +ve and -ve

Single arm

bemcentinib, 200mg daily

+pembrolizumab, 200mg q3wks

Interimanalysis

(n = 24 patients)

Finalanalysis

(n = 48 patients)

• Histopathologically or cytologically documented Stage IV adenocarcinoma of the lung• Has disease progression on or after a prior platinum-containing chemotherapy• Measurable disease as defined by RECIST 1.1• Provision of suitable fresh tumour tissue for the analysis of AXL kinase expression and PD-L1 expression• Eastern Cooperative Oncology Group (ECOG) performance score 0 or 1• Not received more than one prior line of chemotherapy for advanced or metastatic adenocarcinoma of the lung• No prior therapy with an immunomodulatory agent • No known active central nervous system (CNS) metastases and/or carcinomatous meningitis• No recent or ongoing systemic steroid therapy

NCT03184571: Phase II clinical trial of selective AXL inhibitor bemcentinib in combination withpembrolizumab

Simon-like two stage design enrolling up to 48 patients

Key inclusion and exclusion criteria Assessments - efficacy & safety• Response was assessed every 9 weeks per RECIST v1.1• Adverse events were assessed by CTCAE v4.03• Evaluable: ≥1 dose of study treatment as of data cutoff (23 Apr 2019)

Biomarker analysis• PD-L1 and AXL expression by IHC• Soluble protein biomarkers by liquid biopsy• Immune cell populations by multi-spectral imaging

Endpoints• Primary: ORR• Secondary: DCR, DoR, TtP, Survival at 12 months, response by biomarker expression

Study rationale

Anti PD-1 therapies in second line metastatic non-small cell lung cancer (NSCLC)

AXL receptor tyrosine kinase and selective AXL inhibitor bemcentinib

• Pembrolizumab as a single agent is indicated for the treatment of patients with metastatic NSCLC with disease progression on or after platinum-containing chemotherapy1,2

• KEYNOTE-0015, a multi-cohort phase I study designed to define and validate expression levels of PD-L1 associated with the likelihood of clinical benefit, showed that pembrolizumab monotherapy efficacy is correlated with PD-L1 levels • Novel combination treatment strategies are needed to improve efficacy of pembrolizumab while limiting additive toxicity

• AXL is a receptor tyrosine kinase expressed on tumour and immune cells and a member of the TAM family (Tyro-AXL-Mer) of kinases• AXL is overexpressed in response to a hostile tumour micro-environment and drives a tumour survival programme including immune escape, anti-tumour therapy resistance & metastasis3

• AXL is a negative prognostic factor in a multitude of cancers including NSCLC4

• Bemcentinib, a first-in-class highly selective, potent, and orally bioavailable inhibitor of AXL, has been shown to potently improve the efficacy of checkpoint blockade in murine pre-clinical models of NSCLC

n PR SD PD ORR (%)

CBR (%)

Overall** 35 10 12 13 29% 63%

AXL 28

Positive* 15 6 4 5 40% 67%

Negative 13 2 5 6 15% 54%

PD-L1 29

PD-L1 strong positive (TPS ≥50%) 2 1 0 1 50% 50%

PD-L1 positive (TPS 1-49%) 12 3 4 5 25% 58%

PD-L1 negative (TPS <1%) 15 4 5 6 27% 60%

*Any AXL expression as measured by IHC (cut off in development)**All patients with radiological assessments included (n=35)