JAK Pseudokinase Domain Variants Highlight nRTK nsSNPs Identified with Next-Generation

Wednesday, November 8, 2017

Matthew Stein, MD (Hematology/Oncology) Presenting at: IASLC (Int’l Assoc. for the Study of Lung Cancer) Yokohama, Japan-October 16, 2017

Abstract 10429

JAK Pseudokinase Domain Variants Highlight nRTK nsSNPs Identified with Next-Generation Sequencing in NSCLC Patients

Type:      Peer Review  Topic:     02. Biology/Pathology

Authors: Matthew K. Stein, Lindsay K. Morris, Mike G. Martin; West Cancer Center, University of Tennessee Health Science Center, Memphis/United States of America

Background

Non-receptor tyrosine kinase (nRTK) pathways are aberrantly activated in cancer, and mutations in nRTKs have potential therapeutic and prognostic importance. Tumor profiling with next-generation sequencing (NGS)enables a gene’s entire coding sequence to be evaluated, facilitating the identification of novel non-synonymous single nucleotide polymorphisms (nsSNPs) in nRTKs.

Method

 

We searched nsSNPs in 14 nRTKs in the tumors of advanced NSCLC patients (pts) at our institution that received NGS with Caris from 2013-2015. All mutations test-defined as pathogenic (PATH) or nsSNPs labelled variants of undetermined significance (VUS) were included. To classify VUS, nsSNPs underwent PolyPhen-2’s in silico analysis to predict pathogenicity. Any VUS predicted-damaging with PolyPhen-2 we denote pnsSNP. nsSNPs were then classified as occurring within or outside of the tyrosine kinase domain (TKD); JAK1-3 pseudokinase domain (PSKD) lesions were also described.

 

Result

157 NSCLC pts were identified with median age 65 (range 26-85); 51% were male; 65%

Caucasian, 35% African-American. 98 nRTK variants were found (93 nsSNPs and 5 PATHs). 5/5 PATHS were PIK3CA. 31/93 (33%) nsSNPs were pnsSNPs and spread among 30 pts. pnsSNPs were found in 12/14 nRTKs with median 2 (range 0-6). The most frequent were JAK3 (6/20 nsSNPs were pnsSNPs), BTK (5/8), ABL1 (3/12), JAK2 (3/11), CDK12 (3/9) and

JAK1 (3/3). 66% were extra-TKD (28% were pnsSNP), 23% TKD-restricted (44%) and 11% PSKD of JAK1-3 (100%). There were 6 N-lobe PSKD, 3 C-lobe PSKD and 1 C-lobe TKD JAK1-3 pnsSNPs (Table 1) at PSKD-TKD contact sites known to harbor the majority of activating JAK mutations. 6/12 JAK pnsSNPs were in pts whose tumors were EGFR-/KRAS-/ALK-/ROS-

/PDL1-.

Table 1: JAK1-3 pnsSNPs in NSCLC patients.

 

 

JAK

 

VUS;

allele frequency

 

Location

Accession Number; Minor allele frequency

(ExAC)

 

Histology

 

Age, race, gender

Genomics (EGFR, KRAS, ALK or ROS1-

rearranged,

PDL1 (%))

 

JAK1

D660N;

66%

PSKD;

N-lobe

rs368904859; T=2.0e-5 Adeno- carcinoma 66, C, M  

Negative

 

 

 

P674S;

9%

PSKD;

N-lobe

 

None

 

Squamous

76, C, M  

PDL1+ (5%)

D739N;

47%

PSKD;

N-lobe

rs759709239;

T=3.3e-5

Large cell 43, C,

M

KRAS+
JAK2 E621D;

30%

PSKD;

N-lobe

None Unspecified 65,

AA, M

Negative
D686H;

13%

PSKD;

N-lobe

None Adeno-

carcinoma

55, C,

M

Negative
C1105F;

41%

TKD;

C-lobe

None Adeno-

carcinoma

73, C,

F

KRAS+, ROS1-

rearranged

JAK3 V55E; 13% FERM None Adeno- carcinoma 74, C, F Negative
Y105H;

21%

FERM None Squamous 68, C,

F

PDL1+ (20%)
R537Q;

47%

PSKD;

N-lobe

rs587778413;

T=4.1e-5

Adeno-

carcinoma

60, C,

F

PDL1+ (65%)
L702P;

53%

PSKD;

C-lobe

rs772117537;

G=1.7e-5

Squamous 80, C,

M

Negative
P745L;

50%

PSKD;

C-lobe

rs776106625;

A=8.3e-6

Adeno-

carcinoma

68, C,

M

EGFR+

(E746_A750del)

L788I; 7% PSKD;

C-lobe

None Squamous 68,

AA, M

Negative

 

Conclusion

>19% NSCLC pts held a pnsSNP with 77% occurring outside of the TKD-proper. The majority of JAK1-3 pnsSNPs localized to the PSKD; their frequency and functional impact should be examined on a larger scale.