Virtual Library
Start Your Search
Pimpin Incharoen
Author of
-
+
P2.01 - Advanced NSCLC (Not CME Accredited Session) (ID 950)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/25/2018, 16:45 - 18:00, Exhibit Hall
-
+
P2.01-90 - PD-L1 Expression as a Predictive Biomarker in Advanced Non-Small Cell Lung Cancer Patients with or without EGFR Mutation (ID 13528)
16:45 - 18:00 | Author(s): Pimpin Incharoen
- Abstract
Background
The prognostic value of PD-L1 expression and its clinical relevance of NSCLC is controversy. The impact of PD-L1 expression as the predictive biomarker for EGFR-TKIs treatment is needed to explore.
a9ded1e5ce5d75814730bb4caaf49419 Method
Medical records of metastatic NSCLC during September 2015-2016 had been reviewed. PD-L1 immunohistochemistry (IHC) staining with Antibody clone 22C3 was used. The PD-L1 positive was defined by tumor proportion score (TPS) > 1%.
4c3880bb027f159e801041b1021e88e8 Result
204 patients were included. Patients with positive PD-L1 expression had significantly increased numbers of metastatic sites (P=0.009) and lung metastasis (P=0.045) compared to PD-L1 negative patients. Overall survival (OS) was longer in PD-L1 negative patients (22.7 months) compared to PD-L1 positive groups (13.6 months) (HR=1.48; P=0.03). Median OS were significantly different with the number of 7.2, 11.1, 25.7, 42.6 months in EGFR-/PD-L1+, EGFR-/PD-L1-, EGFR+/PD-L1+ and EGFR+/PD-L1- , respectively (P<0.001). Among EGFR positive patients, mOS of T790M-/PD-L1+, T790M-/PD-L1-, T790M+/PD-L1-, and T790M+/PD-L1+ were 22.1, 28, 42.6 and 48.4 months, respectively (P=0.03).
8eea62084ca7e541d918e823422bd82e ConclusionPatient characteristics categorized by PD-L1 expression (N = 204) Characteristics PD-L1 Negative
N=134 (65.69%)
PD-L1 Positive
N=70 (34.31%)
P value Age
- Median age (range)
- < 65 years
- > 65 years
65 (36-85)
64 (47.76)
70 (52.24)
65 (35-86)
33 (47.14
37 (52.86)
0.933 Sex
- Male
- Female
62 (46.27)
72 )53.73)
37 (52.86)
33 (47.14)
0.371 ECOG PS
- 0-1
- > 2
116 (86.57)
18 (13.43)
57 (82.61
12 (17.39)
0.452 Smoking status
- Never
- Ex-smoker
- Current smoker
82 (61.65)
36 (27.07)
15 (11.28)
36 (52.17)
18 (26.09)
15 (21.74)
0.131 Mean smoking pack-year (range) 29.72 (2-100) 25.68 (2-40) 0.873 Initial staging
- Recurrent
- Denovo metastasis
32 (23.88)
102 (76.12)
14 (20)
56 (80)
0.529 Histology
- Adenocarcinoma
- Squamous cell carcinoma
- Adenosquamous carcinoma
- Others
117 (87.31)
1 (0.75)
2 (1.49)
14 (10.45)
58 (84.06)
3 (4.35)
2 (2.9)
6 (8.7)
0.288 EGFR mutation
- Negative
- Positive
47 (35.07)
87 (64.93)
32 (45.71)
38 (54.29)
0.092 Exon 19 deletion
- No
- Yes
87 (64.93)
47 (35.07)
50 (71.43)
20 (28.57)
0.348 L858R
- No
- Yes
100 (74.63)
34 (25.37)
53 (75.71)
17 (24.29)
0.865 ALK results
- Negative
- Positive
79 (92.94)
6 (7.06)
47 (97.92)
1 (2.08)
0.421 Number of site of metastasis
- 0-1
- > 2
88 (65.67)
46 (34.33)
37 (52.86)
33 (47.14)
0.009 Lung metastasis
- No
- Yes
92 (69.17)
41 (30.83)
38 (54.29)
32 (45.71)
0.045 Bone metastasis
- No
- Yes
101 (75.37)
33 (24.63)
53 (75.71)
17 (24.29)
0.957 Liver metastasis
- No
- Yes
122 (91.04)
12 (8.96)
62 (88.57)
8 (11.43)
0.573 Pleural metastasis
- No
- Yes
91 (67.91)
43 (32.09)
62 (88.57)
20 (28.57)
0.606 Brain metastasis
- No
- Yes
117 (87.31)
17 (12.69)
58 (82.86)
12 (17.14)
0.387 Adrenal metastasis
- No
- Yes
122 (91.04)
12 (8.96)
62 (88.57)
8 (11.43)
0.573
PD-L1 expression was associated with poorer survival outcomes among advanced NSCLC patients regardless of EGFR mutation status. PD-L1 expression is also the potential of predictive biomarker for EGFR TKIs treatment. The larger studies are needed to identify the prognostic and predictive values in T790M mutation population.
6f8b794f3246b0c1e1780bb4d4d5dc53
-
+
P3.09 - Pathology (Not CME Accredited Session) (ID 975)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 3
- Moderators:
- Coordinates: 9/26/2018, 12:00 - 13:30, Exhibit Hall
-
+
P3.09-08 - Tumor Heterogeneity and Molecular Profile of NSCLC in Thai Population (ID 14016)
12:00 - 13:30 | Author(s): Pimpin Incharoen
- Abstract
Background
Oncogenic driven mutation is the key to develop targeted therapy in lung cancer. Different ethnicity and tumor heterogeneity affect the prevalence of molecular alteration. This study aimed to explore the unique molecular profile of lung adenocarcinoma in Thai population.
a9ded1e5ce5d75814730bb4caaf49419 Method
We studied 166 lung adenocarcinoma patients’ molecular profile using Next Generation Sequencing (NGS) on 45 genes lung cancer panel (Ion Torrent system). Variants from NGS with coverage of higher than 1000X and cut off at 2% alternate variant frequency were considered positive. We validated the positive mutation of EGFR, BRAF, and KRAS by Real- Time PCR using the Amoy DX kit.
4c3880bb027f159e801041b1021e88e8 Result
This study found 68%(113/166) of EGFR mutation, 9.6%(16/166) of BRAF V600E, 32.5% (54/166) of KRAS mutation, 9%(15/166) of MET exon14 splice site, 4.8% (8/166) AKT mutation (E17K), 2.4% (4/166) of ROS1 mutation, 0.6% (1/166) of PIK3CA mutation (H1047R), and 0.6% (1/166) of PTEN mutation. Furthermore, we also found 40 patients (24.1%) who had more than one mutation in each person. We further validated the positive results by Real-Time PCR. Thirteen patients were obtained tissue from different organs and some with different period of time. T790M usually develop later in EGFR-positive patients who failed 1st or 2nd generation EGFR-TKI. Two patients (patient 5&9) who had lung surgery different lobe in same operation, had different mutation in tissues and one patient (patient 13) who obtained tissue from lung and pleural effusion cell block in different period of time had totally different mutation (Table1).
Table1 Tumor heterogeneity profile in lung cancer patients
case
Hetrogenety in different organ
date obtained tissue
Gene mutation
EGFR
KRAS
ROS
PTEN
AKT
MET
BRAF
1
RLL lobectomy
12-Jun-2012
Exon 19 Deletion
negative
negative
negative
negative
negative
negative
Right lung
4-May-2016
Exon 19 Deletion T790M
negative
negative
negative
negative
negative
negative
2
lymph node
16-Mar-2017
negative
negative
negative
negative
negative
negative
negative
Bone
9-Apr-2017
negative
negative
negative
negative
negative
negative
negative
3
Right upper lung biopsy
20-Jan-2016
Exon 19 Deletion
negative
negative
negative
negative
negative
negative
Lung biopsy
31-Mar-2017
T790M
negative
negative
negative
E17K
negative
negative
4
lymph node
13-Jan-2016
negative
G12A
negative
negative
negative
negative
negative
skin
17-Jan-2016
negative
G12V G12D
negative
negative
negative
negative
negative
left humerous
30-Jun-2016
negative
G12V
negative
negative
negative
negative
negative
5
RML lobectomy
13-Nov-2013
Exon 19 Deletion
G13D
negative
negative
negative
negative
negative
LUL lobectomy
14-May-2014
negative
G12D G13D
negative
R233*
negative
negative
negative
LUL lobectomy
14-May-2014
negative
G12V G12D
D2033N
negative
negative
negative
negative
6
Right pleura biopsy
11-Jan-2016
Exon 19 Deletion
negative
negative
negative
negative
negative
negative
RUL biopsy
1-Mar-2017
Exon 19 Deletion
negative
negative
negative
negative
negative
negative
7
RUL biopsy
28-Sep-2015
L858R
negative
negative
negative
negative
negative
negative
Left pleural fluid cell block
27-Jun-2016
negative
negative
negative
negative
negative
negative
negative
8
Right pleural fluid cell block
25-Mar-2016
Exon 19 Deletion
negative
negative
negative
negative
negative
negative
Left pleural fluid cell block
9-Dec-2016
T790M
negative
negative
negative
negative
negative
negative
9
RML lobectomy
14-Mar-2013
negative
G13S
negative
negative
E17K
negative
negative
RLL wedge resectionRLL lo
30-Mar-2017
G719A L861Q
negative
negative
negative
negative
negative
negative
10
RLL lobectomy
26-Aug-2013
negative
G12C G12D
negative
negative
negative
negative
negative
Left lingular lobe segmental resection
9-Oct-2014
negative
G12D
negative
negative
negative
negative
negative
LUL wedge resection
11-Feb-2016
negative
G12D
negative
negative
negative
negative
negative
LUL lobectomy
4-Jun-2017
negative
G12D
negative
negative
negative
negative
negative
LLL resection
9-Oct-2014
negative
G12D
negative
negative
negative
negative
negative
11
Right pleural cell block
21-Jul-2014
L858R
negative
negative
negative
negative
c.3028G>A exon 14 splicing
negative
Ascites
9-Jun-2017
T790M L858R
negative
negative
negative
negative
negative
negative
12
LUL biopsy
24-Feb-2015
L858R
negative
negative
negative
negative
negative
V600E
Lt lung biopsy
8-Jul-2016
L858R
negative
negative
negative
negative
negative
negative
13
RLL biopsy
14-Oct-2015
negative
negative
negative
negative
negative
c.3028+1G>T exon 14 splicing
negative
pleural fluid cell block
15-Nov-2016
Exon 19 Deletion T790M
negative
negative
negative
negative
negative
negative
Thai populations have unique molecular alteration compared to the other ethnicities, especially, higher of BRAF V600E and MET exon14 splice site. Our population also has high co-mutation prevalence. Tumor heterogeneity is needed to explore in the larger cohort.
6f8b794f3246b0c1e1780bb4d4d5dc53 -
+
P3.09-10 - Circulating Cell-Free DNA (cfDNA) Molecular Profile of Thai NSCLC Patients Using Difference Variant Frequency of NGS (ID 13744)
12:00 - 13:30 | Author(s): Pimpin Incharoen
- Abstract
Background
Detecting cfDNA in the liquid biopsy has become a promising method to explore the genetic landscape of tumor heterogeneity. We developed a pilot-study to find the suitable cutoff of variant frequencies detected from liquid biopsy by NGS to track tumor-associated mutations in NSCLC patients.
a9ded1e5ce5d75814730bb4caaf49419 Method
Ninety-four samples (24 early-stage NSCLC, 70 late-stage NSCLC) were collected from Ramathibodi Hospital, Thailand. Profiling cfDNA using Ion Proton NGS platform. Overall average base coverage depth from NGS was 10,000x, all variants selected have read depths >10x in order to reach 0.1% sensitivity. Each of selected variants has threshold variant quality (QUAL) >20. Droplet digital PCR (ddPCR) was performed for EGFR-mutation testing to determine the appropriated cutoff variant frequency from NGS.
4c3880bb027f159e801041b1021e88e8 Result
In early-stage NSCLC, a minimum-threshold variant frequencies at 0.1% could detect EGFR exon19 deletion in all samples (24,100%), with BRAF (12,50%), KRAS (21,87.5%) and other mutations in AKT1, MET, PIK3CA, PTEN, ROS1 (14,58%). None of these mutations identified when using conventional level cutoff at 3% (Table1). ddPCR observed EGFR-mutations in 2 early-stage cases only (8.3%). In late-stage NSCLC, 64 (91.4%) cases were observed multiple mutations, suggesting tumor heterogeneity. At 0.1% cutoff in NGS, Thirty-six (52.9%) cases of EGFR-mutations in NGS and ddPCR were identical. Thirteen (18.6%) samples shown partial discrepancies in the mutations. Interestingly, NGS found EGFR-mutations in 20 (28.6%) samples which ddPCR failed to detect, 12 of them contained T790M. Only one sample (1.4%) using 0.1% cutoff was unable to detect EGFR-mutation. Higher variant allele frequencies were found in EGFR-positive detected by ddPCR compared to not-detected by ddPCR.
Table1 Mutations detected variant allele frequencies detected from liquid biopsy by NGS at minimum threshold cut-off 0.1% variant allele frequencies detected from liquid biopsy by NGS at minimum threshold cut-off 0.1% variant allele frequencies detected from liquid biopsy by NGS at conventional level detection of somatic variants cut-off 3% variant allele frequencies detected from liquid biopsy by NGS at conventional level detection of somatic variants cut-off 3% variant allele frequencies detected from liquid biopsy by ddPCR (EGFR only) variant allele frequencies detected from liquid biopsy by ddPCR (EGFR only) N (%) median (range) N (%) median (range) N (%) median (range) BRAF (V600E) early stage total N=24 12 (50%) 0.8 (0.3-2.5) 0 (0%) 0 NA NA BRAF (V600E) late stage total N=70 11 (15.7%) 0.6 (0.1-1.1) 0 (0%) 0 NA NA KRAS early stage 21 (87.5%) 0.1 (0.1-0.5) 0 (0%) 0 NA NA KRAS late stage 50 (71.4%) 0.2
(0.1-13.6)
3 (4.3%) 11.1 (5.8-14.3) NA NA AKT1 (E17K) early stage 7 (29.2%) 0.1 (0.1-0.7) 0 (0%) 0 NA NA AKT1 (E17K) late stage 12 (17.1%) 0.1 (0.1-0.7) 0 (0%) 0 NA NA MET exon 14 splicing early stage 4 (16.7%)
0.1 (0.1-0.1) 0 (0%) 0 NA NA MET exon 14 splicing late stage 3 (4.3%) 0.2 (0.2-0.2) 0 (0%) 0 NA NA PIK3CA early stage 9 (37.5%) 0.2 (0.1-0.8) 0 (0%) 0 NA NA PIK3CA late stage 19 (27.1%) 0.3 (0.1-0.7) 0 (0%) 0 NA NA PTEN (R233*) early stage 6 (25.0%) 0.1 (0.1-0.4) 0 (0%) 0 NA NA PTEN (R233*) late stage 11 (15.7%) 0.1 (0.1-0.2) 0 (0%) 0 NA NA ROS1
early stage 0 (0%) 0 0 (0%) 0 NA NA ROS1
late stage 4 (5.7%) 0.55
(0.1-0.7)
0 (0%) 0 NA NA EGFR
Exon 19 Deletion
early stage 24 (100%) 0.35
(0.1-2.1)
0 (0%) 0 1 (4.2%) 0.5 (0.5-0.5) EGFR
Exon 19 Deletion
late stage 25 (35.7%) 0.6
(0.1-49.0)
6 (8.6%) 9.4 (4.5-49.5) 20 (28.6%) 0.65 (0-49.0) EGFR L858R early stage 5 (20.8%) 0.2 (0.1-0.5) 0 (0%) 0 0 (0%) 0 EGFR L858R late stage 21 (30%) 1.4 (0.1-9.7) 4 (5.7%) 4.6 (4.4-6.4) 14 (20%) 1.8 (0.3-9.7) EGFR T790M early stage 8 (33.3%) 0.1 (0.1-0.2) 0 (0%) 0 0 (0%) 0 EGFR T790M late stage 30 (42.9%) 0.1 (0.1-4.6) 2 (2.9%) 7.5 (5.3-9.7) 16 (22.9%) 0.15 (0-4.1) EGFR Exon18 (G719X) early stage 6 (25%) 0.2 (0.1-0.4) 0 (0%) 0 1 (4.2%) 0.4 (0.4-0.4) EGFR Exon18 (G719X) late stage 4 (5.7%) 4.5
(0.1-49.2)
2 (2.9%) 27.9
(6.7-49.2)
2 (2.9%) 25.6 (2-49.2) EGFR Exon 20 Insertion early stage 0 (0%) 0 0 (0%) 0 0 (0%) 0 EGFR Exon 20 Insertion late stage 1 (1.4%) 73.8
(73.8-73.8)
1 (1.4%) 91.7
(91.7-91.7)
0 (0%) 0
Detecting variant frequencies at 0.1% could reveal more hidden tumor-associated mutations compared to variant frequency cutoff at 3%. With a careful validation, profiling cfDNA using NGS can be a crucial method to accurately select treatment for NSCLC patients in the future.
6f8b794f3246b0c1e1780bb4d4d5dc53 -
+
P3.09-19 - Matched Thai Lung Cancer Patients Tissue and cfDNA Molecular Profile by NGS (ID 14341)
12:00 - 13:30 | Author(s): Pimpin Incharoen
- Abstract
Background
Liquid biopsy is the new non-invasive technology to explore the molecular profile. We evaluate molecular alteration from matched tissue and liquid specimen in NSCLC patients using NGS.
a9ded1e5ce5d75814730bb4caaf49419 Method
A total 61 matched tumors and cfDNA in NSCLC patients were retrieved for DNA extraction. All qualified samples were analyzed by using Next Generation Sequencing (NGS) with Gene read Qiagen Lung Cancer Panel sequencing 45 Genes on Ion Torrent system.Variants from NGS with coverage of higher than 1000X of tissues and 10000X of liquid biopsy, cutoff at 3% variant frequency were considered positive. Each detected mutation was validated by the different method. EGFR-mutation detected at this cutoff was validated by Real-time PCR technique using the ARMS-PCR (Amoy DX Kit) for tissue samples and droplet-digital PCR (ddPCR) for blood samples in all samples.
4c3880bb027f159e801041b1021e88e8 Result
This study found 59.0% and 19.7% of EGFR mutation, 14.75% and 8.20% of KRAS mutation in tissues and cfDNA, respectively. Moreover, we found 3.29% of BRAF V600E, 1.64% of MET exon14 splice site, and 1.64% of ROS1 mutation in tissue NGS and also confirmed by the other techniques, but there was none of these mutations in the blood sample. Looking at EGFR-mutation detected by different techniques, higher sensitivity, specificity, positive-predictive value, negative-predictive values, and concordant rate were found in tissue NGS and liquid ddPCR compared with liquid NGS at cutoff 3% of variant frequency detection, when the gold standard for validation was ARMS-PCR in tissue testing (Table1).
Table1. Performance of NGS, ARMS and ddPCR for EGFR mutation detection in matched tissue and cfDNA in lung cancer patients. Tissue by ARMS PCR
Result
Tissues by NGS
negative
positive
total
Sensitivity= 87.7%
negative
14
11
25
Specificity= 100%
positive
0
36
36
Positive predictive value= 100%
Total
14
47
61
Negative predictive value= 75.6%
Concordance = 82%
Tissue by ARMS PCR
cfDNA by NGS
negative
positive
total
Sensitivity= 35.7%
negative
12
37
49
Specificity= 98.2%
positive
2
10
12
Positive predictive value= 97.9%
Total
14
47
61
Negative predictive value= 38.9%
Concordance = 42.6%
Tissue by ARMS PCR
cfDNA by ddPCR
negative
positive
total
Sensitivity= 82.7%
negative
14
14
28
Specificity= 100%
positive
0
33
33
Positive predictive value= 100%
Total
14
47
61
Negative predictive value= 69.4%
Concordance = 77.05%
Tissue by NGS
cfDNA by NGS
negative
positive
total
Sensitivity= 36%
negative
20
29
49
Specificity= 93.2%
positive
5
7
12
Positive predictive value= 84.8%
Total
25
36
61
Negative predictive value= 55.8%
Concordance = 60.66%
cfDNA by ddPCR
cfDNA by NGS
negative
positive
total
Sensitivity= 45.5%
negative
25
24
49
Specificity= 97.7%
positive
3
9
12
Positive predictive value= 94.5%
Total
28
33
61
Negative predictive value= 65.6%
Concordance = 65.58%
Using tissue for molecular profile testing is still being the gold standard testing. Liquid biopsy is less invasive, but in our study, liquid NGS performed less sensitivity, specificity, PPV, NPV, and concordance rate compared with tissue NGS. We need to explore more for proper cutoff of variant allele frequency to develop more sensitivity and specificity of liquid NGS.
6f8b794f3246b0c1e1780bb4d4d5dc53
-
+
P3.15 - Treatment in the Real World - Support, Survivorship, Systems Research (Not CME Accredited Session) (ID 981)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/26/2018, 12:00 - 13:30, Exhibit Hall
-
+
P3.15-24 - Ramathibodi Lung Cancer Consortium (RLC) Model: Multidisciplinary Team Approach Improves Lung Cancer Patients’ Survival Outcome (ID 14396)
12:00 - 13:30 | Author(s): Pimpin Incharoen
- Abstract
Background
Technologies for investigation, diagnosis, and treatment for lung cancer are advance to improve patients’ survival. Many investigations will be performed once the patients were suspected to have lung cancer. All processes sometime take a long time to complete investigations before starting the treatment which may affect treatment outcomes.
a9ded1e5ce5d75814730bb4caaf49419 Method
Ramathibodi Lung Cancer Consortium (RLC) was established in October 2014, aims to help the patients accessing all investigations and treatments faster by multidisciplinary team (MDT) approach and patient-center with one-stop service system. We conduct RLC meeting every 1st and 3rd Tuesday of the month. As of May 2017, 200 new lung cancer patients were solved all problems by RLC team. We collected and analyzed the data of lung cancer patients between 2 groups. The first group was the patient whom underwent RLC model (160 cases) and the second group was control group which was the patient whom diagnosed before establishing RLC (72 cases). Our primary endpoint was time from first visit to first treatment. Secondary endpoints were time from first visit to first interventions, number visits from first visit to first treatment, and overall survival (OS). We also did subgroups analysis for time from first visit to first biopsy, to first imaging study, to surgery, to chemotherapy, and to radiation.
4c3880bb027f159e801041b1021e88e8 Result
Median time from first visit to first treatment was significantly decreased in RLC group (14 days) compared to 57 days in control group with HR of 2.86 (95% CI; 2.11-3.87, P<0.001). Median time from first visit to first intervention was also significantly decreased in RLC group (2 days) compared to 11 days in control group with HR of 2.01 (95% CI; 1.53-2.62, P<0.001). Median number of hospital visits was significantly lower in RLC group (1 visits) compared to control group (8 visits). All subgroup analyses showed significantly decreased duration of each investigation and each treatment in RLC group. In survival analysis, lung cancer patients whom underwent RLC model had significantly longer mOS compared to control group, especially in stage 3 and 4 disease [mOS = 2.4 vs 0.8 years, HR=0.42 (95% CI; 0.3-0.7, P<0.001)].
8eea62084ca7e541d918e823422bd82e Conclusion
RLC model is a very useful model helping lung cancer patients to access treatment and investigations in short period of time and translate to have significantly longer survival. RLC model also provides the cooperation in lung cancer research. This model should be applied for all cancers treatment. Working as MDT is the utmost importance for cancer treatment.
6f8b794f3246b0c1e1780bb4d4d5dc53