Prognostic Implications of Multiplex Detection of KRAS Mutations in Cell-Free DNAfrom Patients with Pancreatic Ductal Adenocarcinoma.
KRAS  is one of the most common oncogenes mutated in pancreatic cancer (7), and such mutations have been observed in >90% ofPDAC cases (8, 9). The most frequent mutations of KRAS occur at codon 12, and among these the most common are G12D, G12V, and G12R, with frequencies of 51%, 30%, and 12%, respectively (10). Although the presence ofa KRAS mutation in tumor tissues has been reported to be associated with a decrease in overall survival (OS) (11), the widespread use of KRAS mutations as a prognostic indicator is limited by difficulties in obtaining tissues from patients because only approximately 20% of patients present with a resectable tumor. In this context, a prognostic, noninvasive blood test for PDAC would be very valuable.
Cell-free DNA (cfDNA), a generic term for nucleic acid fragments that exist separate from cells in the blood, is known to include products from cancer cells that circulate in the bloodstream (12). cfDNA is delivered into the blood of cancer patients through active secretion as well as various cellular changes, such as apoptosis, necrosis, and the proliferation of cancer cells; moreover, the volume of cancer tissue can be an important factor in determining the quantity of cfDNA (13).
In this prospective study, we performed multiplex detection of frequent KRAS mutations in cfDNA in plasma and investigated the prognostic implications of KRAS mutation detection in PDAC patients.
Materials and Methods
STUDY DESIGN AND SAMPLE COLLECTION
A total of 106 patients diagnosed with PDAC at the National Cancer Center in Korea between March 2015 and January 2017 were recruited for this study. Patients gave informed consent for involvement in this study, which was approved by our Institutional Review Board (No. NCC2015-0054 and NCC2016-001). Patients were divided into 3 clinical stage groups: patients with surgical resection, patients with locally advanced disease, and patients with metastatic disease. Blood samples were collected before treatment and every 3 months after treatment. In total, 13 primary frozen tumor tissues and 64 formalin-fixed paraffin-embedded (FFPE) tissues were obtained from the biobank of the National Cancer Center, Korea.
EXTRACTION OF cfDNA FROM PLASMA AND GENOMIC DNA FROM PRIMARY TISSUES
Blood samples were processed within 2 h after blood draw to ensure the integrity of cfDNA. cfDNA was extracted from plasma by first centrifuging whole-blood specimens for 10 min at 1800g and 4 [degrees]C, after which the supernatant was centrifuged again (10 min at 16000g and 4 [degrees]C) to remove any remaining contaminating cells. cfDNA was extracted from 1 mL of plasma with a QIAamp Circulating Nucleic Acid Kit (Qiagen) according to the manufacturer's instructions and quantified on a Qubit 2.0 fluorometer using a Qubit dsDNA HS (High Sensitivity) Assay Kit (Life Technologies). Genomic DNA from primary tumor tissues and FFPE tissues was extracted with a tissue DNA Kit and GeneRead FFPE DNA Kit (Qiagen).
DETECTION OF KRAS MUTATIONS IN cfDNA AND TISSUE DNA BY DROPLET DIGITAL PCR
KRAS mutations in cfDNA extracted from plasma and genomic DNA extracted from tissues were detected by droplet digital PCR (ddPCR) on a QX200 Droplet Digital PCR System (Bio-Rad Laboratories) using a KRAS screening multiplex ddPCR Kit, which covers 7 common KRAS mutations (G12A, G12C, G12D, G12R, G12S, G12V, and G13D). Analyses were performed by QuantaSoft software (Bio-Rad).
The limit of detection for KRAS mutations was evaluated by serially diluting cfDNA from the mutated pancreatic cancer cell line, CFPAC-1, producing a standard curve with a low-end fractional abundance of KRAS mutations of 0.01%. The limit of detection was defined as 0.1%; thus, samples with a fractional abundance >0.1% were considered positive for KRAS mutation.
MEASUREMENT OF CA19-9 AND CEA
Serum concentrations of CA19-9 were measured by use of a CA19-9 RIA kit (Centocor), which has a recommended upper reference limit of 37 U/mL. CEA concentrations were measured by chemiluminescent microparticle immunoassay with an Architect i1000SR Immunoassay Analyzer (Abbott Laboratories). The cutoff concentrations for CA19-9 and CEA were determined based on known values of 37 U/mL and 5 U/mL, respectively.
The CA19-9 and CEA were dichotomized with known cutoff concentrations, and cfDNA concentration, KRAS mutation concentration, and KRAS fractional abundance were dichotomized based on the median value of the total dataset. Associations between clinical factors, including age, sex, stage, Eastern Cooperative Oncology Group score, tumor location, initial treatment, CA19-9, CEA, cfDNA concentration, KRAS mutation concentration, KRAS fractional abundance, and patients' survival, were analyzed with univariable or multivariable Cox proportionalhazards regression models. After the backward variable selection with an elimination criterion of P value of >0.05 was performed, only the stage was adjusted in progression-free survival (PFS) and OS multivariable models. OS was calculated from the day of diagnosis to the day of last follow-up or death. PFS was measured from the day of diagnosis to the day of progression or death. The effects of cfDNA concentration and KRAS mutation on OS or PFS were presented as hazard ratios (HRs) and 95% CIs. Survival curves were estimated with use of the Kaplan--Meier method, and the survival difference was tested by use of the log-rank test. Comparison of changes in cfDNA concentration and KRAS fractional abundance after 6 months was tested with a Wilcoxon rank-sum test, Pearson [chi square] test, or Fisher exact test for each variable, as appropriate. The ability of factors to predict 3-month and 6-month OS was assessed with the receiver operating characteristics (ROC) curve and area under the ROC curve (AUC). A P value of <0.05 was considered statistically significant. All statistical analyses were performed with use of SAS (version 9.4; SAS Institute Inc) and R statistical software (version 3.3.2).
CHARACTERISTICS OF PANCREATIC CANCER PATIENTS
The characteristics of the 106 patients included in the study are summarized in Table 1. Of the 106 patients enrolled, 67 were men, median age was 66 years, and the median follow-up period was 10.03 months (range, 0.07--19.96 months). Patients with resectable, locally advanced, and metastatic cancers accounted for 38.7% (n = 41), 23.6% (n = 25), and 37.7% (n = 40) of all patients, respectively. The resectable group had a significantly lower HR than locally advanced and metastatic groups in PFS (HR, 2.73; 95% CI, 1.34-5.59; P = 0.006 and HR, 4.23; 95% CI, 2.19-8.17; P < 0.001) and OS (HR, 5.55; 95% CI, 2.18-14.10; P< 0.001 and HR, 8.94; 95% CI, 3.64-21.97; P < 0.001).
CLINICAL OUTCOME ACCORDING TO KRAS MUTATION CONCENTRATION AND FRACTIONAL ABUNDANCE
The KRAS mutation concentration and abundance were significantly higher in the metastatic group than in the resectable and locally advanced groups (Fig. 1A). KRAS mutations were detected in 96.1% (74 out of 77) and 77.9% (60 out of 77) of the tissue and cfDNA, respectively. The concordance of KRAS mutations between tumor tissue DNA and cfDNA was 75.3% (Fig. 1B). Although the mutation in the tissues was not completely consistent with cfDNA, the concordance rate of KRAS mutation between tissue and cfDNA was 76.6% (59 out of 77; see Table 1 in the Data Supplement that accompanies the online version of this article at http://www. clinchem.org/content/vol64/issue4). Sensitivity, specificity, accuracy, negative predictive value, and positive predictive value for data comparing the KRAS mutation fractional abundance in tumor tissues and cfDNA according to the detection limit of 0.1% were 78.4, 33.3, 76.6, 96.7, and 5.9, respectively. The rate of KRAS mutations in cfDNA was higher in the metastatic group (n = 31, 86.1%) than in resectable (n = 24, 68.6%) or locally advanced (n = 5, 83.3%) groups. Median cfDNA concentration, KRAS mutation concentration, and fractional abundance were 427 ng/mL, 0.165 copies/[micro]L, and 0.415%, respectively.
Univariable analyses demonstrated that CEA concentration (HR, 1.70; 95% CI, 1.03-2.82), KRAS mutation concentration (HR, 2.64; 95% CI, 1.55-4.50), and KRAS fraction (HR, 2.18; 95% CI, 1.30-3.66) were significant factors for PFS, whereas CA19-9 (HR, 2.03; 95% CI, 1.10-3.75), KRAS mutation concentration (HR, 2.54; 95% CI, 1.41-4.56), and KRAS fraction (HR, 2.19; 95% CI, 1.23-3.90) were significant factors for OS. Multivariable analyses demonstrated that KRAS mutation concentration (HR, 2.08; 95% CI, 1.20-3.63) and KRAS fraction (HR, 1.73; 95% CI, 1.02-2.95) of cfDNA were significantly associated with PFS. KRAS mutation concentration (HR, 1.97; 95% CI, 1.05-3.67) was the only significant factor for OS (Table 2 and Fig. 1 in the online Data Supplement).
A subsequent subgroup analysis of survival showed that those with a low KRAS mutation concentration and fractional abundance (low subgroups) showed significantly longer PFS (12.6 vs 4.7 months, P < 0.001; 8.7 vs 5.1 months, P = 0.003) and OS (13.0 vs 8.0 months, P = 0.001; 12.6 vs 8.0 months, P = 0.006) than high subgroups. (Fig. 2A, B). Among patients with resectable disease, CA19-9, CEA, and cfDNA concentrations were not significantly different between high and low subgroups. PFS was significantly lower in subgroups with a high KRAS mutation concentration (P = 0.016) and fractional abundance (P = 0.02) than in low subgroups (Fig. 2C, D). In locally advanced and metastatic groups, high and low KRAS mutation concentration and KRAS fractional abundance showed no association with PFS or OS (Fig. 2 in the online Data Supplement).
An analysis of changes in biomarker levels before and after treatment (3 months, n = 56; 6 months, n = 46) showed that OS was significantly lower in the group with increased cfDNA concentration, KRAS mutation concentration, and fractional abundance values 6 months after treatment compared with before-treatment values (P < 0.001, P = 0.013 and P = 0.036, respectively; see Fig. 3 and Table 2 in the online Data Supplement). Neither PFS nor OS differed between 3 months after treatment and pretreatment.
ADDITIVE BENEFIT OF CA19-9 CONCENTRATIONS AND KRAS MUTATION STATUS AS PROGNOSTIC BIOMARKERS
AUC values of combined CA19-9 concentrations and KRAS mutation concentration at 3 months and 6 months were 0.746 and 0.705, respectively, and the corresponding AUC values for fractional abundance at 3 months and 6 months were 0.729 and 0.694, respectively. These values are significantly higher than those calculated with CA19-9 alone (0.590 and 0.595), suggesting that when combined with the cancer biomarker CA19-9, the KRAS mutation concentration showed additive benefit for the prediction of OS (see Table 3 in the online Data Supplement).
The main clinical applications of cfDNA include the detection of cancer, prediction of prognosis, and monitoring of systemic therapies (14). Tumor heterogeneity also offers an important rationale for blood-based analysis compared with analysis of a single tissue. This study showed that KRAS mutant concentration (copies/[micro]L) and fractional abundance (%) in plasma cfDNA were associated with prognosis in a prospective cohort of PDAC patients. These findings suggest that cfDNA can serve as a biomarker to aid in determining which tumors will recur and identifying patients who would benefit from adjuvant chemotherapy after surgery. PDAC patients with a higher KRAS mutation concentration at diagnosis who were treated with resection showed poor prognosis in terms of PFS (P = 0.016) and OS (P = 0.072), and patients who had increased KRAS mutation fractional abundance values at 6 months had a shorter OS (P = 0.036) than those who had decreased values.
Previous reports on cfDNA in pancreatic cancer (15--25) are summarized in Appendix Table 1. The current study showed a higher rate of KRAS mutation in cfDNA (76.4%) during all stages of PDAC than a previous study using the same ddPCR method, which reported that 26% of cfDNA samples were positive for KRAS mutations (18). This difference might reflect the fact that our study used a multiplex KRAS mutation detection kit that encompasses 7 common mutations (G12A, G12C, G12D, G12R, G12S, G12V, and G13D), whereas the previous study investigated 3 KRAS mutation sites (G12D, G12V, and G12R). Another reason for the difference may be that the cutoff criteria used in each study are different. The lack of a criterion to judge mutations as positive and differing limits of detection among the test methods to detect the mutation also may have produced differences in the positive rates.
Several studies have reported prognostic implications of the fractional abundance of KRAS mutations among all stage groups and in the metastatic group (15, 17). In one study by Kinugasa et al. (15), results based on an analysis of the specific KRAS mutations, G12D, G12V, and G12R, were similar to ours, as reflected in total HR value (1.84). Our study has the advantage of being able to test 7 positions simultaneously. The study ofTakai et al. (17) also investigated prognostic implications, but their picoliter ddPCR method was different from ours, and their reported mutation detection rate (32%) was very different from ours.
Comparing the methods used to detect KRAS mutation, the most commonly used method, and the one used for our study, is ddPCR, which has yielded reported KRAS mutation-positivity rates ranging from 26% to 62.6%. Other methods, such as peptide-nucleic acid clamp PCR and a scorpion amplification refractory mutation system have yielded positive rates that differ according to the coverage of KRAS sites (16, 19).
Targeting RAS is the most obvious and attractive approach for developing a treatment for PDAC. Despite the disappointing history of anti-RAS drug discovery, recent data have renewed optimism for targeting RAS directly (26). The detection of circulating KRAS mutations in PDAC cancer patients may offer a number of critical advantages for essentially real-time monitoring of tumor responses to anti-RAS therapy in PDAC cancer patients.
In the present study, KRAS mutations in cfDNA proved more effective when used together with the existing biomarkers, CA19-9 and CEA. Furthermore, circulating KRAS mutations may guide therapeutic strategies, especially in patients with resectable PDAC. In conclusion, our findings collectively suggest that KRAS mutations in cfDNA are likely to be of practical value for clinical applications.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
All authors reviewed the manuscript. M.K. Kim conducted experiments. S.M. Woo, W.J. Lee, and S.-S. Han acquired data. B. Park and J. Joo performed statistics. Y.-H. Kim and K.-A. Yoon reviewed experiments. S.-J. Park and S.-Y. Kong designed research.
Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:
Employment or Leadership: None declared.
Consultant or Advisory Role: None declared.
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: The National Cancer Center, Korea (grant no. 1510203) to institution.
Expert Testimony: None declared.
Patents: None declared.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or final approval of manuscript.
Acknowledgment: The authors thank Young Hwa Kang for her help in collecting patient specimens and clinical data.
(1.) Siegel RL, Miller KD, Jemal A. Cancerstatistics, 2017.CA CancerJ Clin 2017;67:7-30.
(2.) Jung KW, Won YJ, Oh CM, Kong HJ, Lee DH, Lee KH, et al. Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2014. Cancer Res Treat 2017; 49:292-305.
(3.) Duan H, Lu J, Lu T, Gao J, Zhang J, Xu Y, et al. Comparison of EGFR mutationstatus between plasmaand tumor tissue in non-small cell lung cancer using the Scorpion ARMS method and the possible prognostic significance of plasma EGFR mutation status. Int J Clin Exp Pathol 2015;8:13136-45.
(4.) Humphris JL, Chang DK, Johns AL, Scarlett CJ, Pajic M, Jones MD, et al. The prognostic and predictive value of serum CA19.9 in pancreatic cancer. Ann Oncol 2012; 23:1713-22.
(5.) Poruk KE, Gay DZ, Brown K, Mulvihill JD, Boucher KM, Scaife CL, et al. The clinical utility of CA 19-9 in pancreatic adenocarcinoma: diagnostic and prognostic updates. Curr Mol Med 2013;13:340-51.
(6.) Swords DS, Firpo MA, Scaife CL, Mulvihill SJ. Biomarkers in pancreatic adenocarcinoma: current perspectives. Onco TargetsTher2016;9:7459-67.
(7.) di Magliano MP, Logsdon CD. Rolesfor KRAS in pancreatic tumor development and progression. Gastroenterology 2013;144:1220-9.
(8.) Collins MA and Pasca di Magliano M. KRAS as a key oncogene and therapeutic target in pancreatic cancer. Front Physiol 2013;4:407.
(9.) Rachakonda PS, Bauer AS, Xie H, Campa D, Rizzato C, Canzian F, et al. Somatic mutations in exocrine pancreatic tumors: association with patient survival. PLoS One 2013;8:e60870.
(10.) Bryant KL, Mancias JD, Kimmelman AC, Der CJ. KRAS: feeding pancreatic cancer proliferation. Trends Biochem Sci 2014;39:91-100.
(11.) Sinn BV, Striefler JK, Rudl MA, Lehmann A, Bahra M, Denkert C, et al. KRAS mutations in codon 12 or 13 are associated with worse prognosis in pancreatic ductal adenocarcinoma. Pancreas 2014;43:578-83.
(12.) Pantel K, Alix-Panabieres C. Real-time liquid biopsy in cancer patients: fact or fiction? Cancer Res 2013;73: 6384-8.
(13.) Schwarzenbach H, Hoon DS, Pantel K. Cell-free nucleic acids as biomarkers in cancer patients. Nat Rev Cancer 2011;11:426-37.
(14.) Alix-Panabieres C, Pantel K. Clinical applications of circulating tumor cells and circulating tumor DNA as liquid biopsy. Cancer Discov 2016;6:479-91.
(15.) Kinugasa H, Nouso K, Miyahara K, Morimoto Y, Dohi C, Tsutsumi K, et al. Detection of K-ras gene mutation by liquid biopsy in patients with pancreatic cancer. Cancer 2015;121:2271-80.
(16.) Tjensvoll K, Lapin M, Buhl T, Oltedal S, Steen-Ottosen Berry K, Gilje B, et al. Clinical relevance of circulating KRAS mutated DNA in plasma from patients with advanced pancreatic cancer. Mol Oncol 2016;10:635-43.
(17.) Takai E, Totoki Y, Nakamura H, Morizane C, Nara S, Hama N, et al. Clinical utility of circulating tumor DNA for molecular assessment in pancreatic cancer. Sci Rep 2015;5:18425.
(18.) Earl J, Garcia-Nieto S, Martinez-Avila JC, Montans J, Sanjuanbenito A, Rodriguez-Garrote M, et al. Circulating tumor cells (CTC) and KRAS mutant circulating free DNA (cfDNA) detection in peripheral blood as biomarkers in patients diagnosed with exocrine pancreatic can cer. BMC Cancer 2015;15:797.
(19.) Semrad T, Barzi A, Lenz HJ, Hutchins IM, Kim EJ, Gong IY, et al. Pharmacodynamic separation of gemcitabine and erlotinib in locally advanced or metastatic pancreatic cancer: therapeutic and biomarker results. Int J Clin Oncol 2015;20:518-24.
(20.) Zill OA, Greene C, Sebisanovic D, Siew LM, Leng J, Vu M, et al. Cell-free DNA next-generation sequencing in pancreatobiliary carcinomas. Cancer Discov 2015;5:1040-8.
(21.) Hadano N, Murakami Y, Uemura K, Hashimoto Y, Kondo N, Nakagawa N, et al. Prognostic value of circulating tumour DNA in patients undergoing curative resection for pancreatic cancer. Br J Cancer 2016;115: 59-65.
(22.) Le Calvez-Kelm F, Foll M, Wozniak MB, Delhomme TM, Durand G, Chopard P, et al. KRAS mutations in blood circulating cell-free DNA: a pancreatic cancer casecontrol. Oncotarget 2016;7:78827-40.
(23.) Cheng H, Liu C, Jiang J, Luo G, Lu Y, Jin K, et al. Analysis of ctDNA to predict prognosis and monitor treatment responses in metastatic pancreatic cancer patients. Int J Cancer 2017;140:2344 -50.
(24.) Pietrasz D, Pecuchet N, Garlan F, Didelot A, Dubreuil O, Doat S, et al. Plasma circulating tumor DNA in pancreatic cancer patients is a prognostic marker. Clin Cancer Res 2017;23:116-123:116-23.
(25.) FujiiT, Barzi A, Sartore-Bianchi A, Cassingena A, Siravegna G, Karp D, et al. Mutation-enrichment next-generation sequencing for quantitative detection of KRAS mutations in urine cell-free DNA from patients with advanced cancers. Clin Cancer Res 2017;23:3657-66.
(26.) Papke B, Der CJ. Drugging RAS: know the enemy. Science 2017;355:1158-63.
Min Kyeong Kim, [dagger] Sang Myung Woo, [dagger] Boram Park,  Kyong-Ah Yoon,  Yun-Hee Kim, [5,6] Jungnam Joo,  Woo Jin Lee,  Sung-Sik Han,  Sang-Jae Park,  and Sun-Young Kong [1,5,7]*
 Translational Cancer Research Branch, DivisionofTranslational Science, National Cancer Center, Goyang, Korea;  Centerfor Liver Cancer, National Cancer Center, Goyang, Korea;  Biometrics Research Branch, Division of Cancer Epidemiology and Management, National Cancer Center, Goyang, Korea;  College of Veterinary Medicine, Konkuk University, Seoul, Korea;  Cancer Biomedical Science, National Cancer Center Graduate School of Cancer Science and Policy, National Cancer Center, Goyang, Korea;  Molecular Imaging Branch, Division of Convergence Technology, National Cancer Center, Goyang, Korea;  Department of Laboratory Medicine, Centerfor Diagnostic Oncology, National Cancer Center, Goyang, Korea.
* Address correspondence to this author at: National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si, Gyeonggi-do, 10408, Republic of Korea. Fax +82-31-9201268; e-mail email@example.com.
[dagger] Min Kyeong Kim and Sang Myung Woo contributed equally to this work.
Received October 24,2017; accepted January 3,2018.
Previously published online at DOI: 10.1373/clinchem.2017.283721
 Nonstandard abbreviations: CA 19-9, carbohydrate antigen 19-9; CEA, carcinoembryonic antigen; PDAC, pancreatic ductal adenocarcinoma; OS, overall survival; cfDNA, cell-free DNA; FFPE, formalin-fixed paraffin-embedded; ddPCR, droplet digital PCR; PFS, progression-free survival; HR, hazard ratio; ROC, receiver operating characteristics; AUC, area under the curve.
 Human Genes: KRAS, KRAS proto-oncogene, GTPase.
Caption: Fig. 2. Kaplan-Meier curves of progression-free survival (PFS) and overall survival (OS) KRAS mutation concentrations and fractional abundance in the overall combined groups (A and B) and in the resectable group (C and D).
In (A) and (B), KRAS mutation concentrations and fraction abundance showed an association with PFS (P <0.001 and P = 0.001) and OS (P = 0.003 and P = 0.006) in all stages. In (C) and (D), high KRAS mutation concentrations and fractional abundance were associated with worse survival rates (P = 0.016, P = 0.02) for the resectable group.
Caption: Fig. 1. Comparison of KRAS mutation concentrations and KRAS mutation fractional abundances between blood cell-free (cf) DNA and tissue DNA according to stages. Distribution of KRAS mutation concentrations and KRAS mutation fractional abundances in cfDNA according to stage (A). KRAS mutations fractional abundance (%) in tissues and cfDNA were compared in 3 stages as resectable (n = 35), locally advanced (n = 6), and metastatic groups (n=36). The detection rate of KRAS mutations was higher in tissue DNA than in cfDNA (B). Among those groups, the metastatic group was associated with a high number of mutations. The correlation coefficient between cfDNA and tissue DNA was 0.57.
Table 1. Baseline characteristics of pancreatic adenocarcinoma patients. PFS (univariable) No. of Event (%> patients (%) Total number 106 Age Median (range) 66.0 (40.0-88.0) 61 (57.5) Sex Female 39 (36.8) 23(59.0) Male 67 (63.2) 38(56.7) Stage Resectable 41 (38.7) 13(31.7) Locally advanced 25 (23.6) 18(72.0) Metastatic 40 (37.7) 30(75.0) ECOG 0 65 (61.3) 35 (53.9) 1 + 2 41 (38.7) 26(63.4) Tumor location Body and Tail 53 (50.0) 33(62.3) Head and Neck 53 (50.0) 28(52.8) Initial treatment Surgery 36 (34.0) 10(27.8) CCRTorPBT 13 (12.3) 10(76.9) Chemotherapy 33 (31.1) 22 (66.7) Supportive care only 24(22.6) 19(79.2) Operation name (N = 39) DP + TP 13 (33.3) 3(23.1) PD + PPPD 26 (66.7) 8(30.8) Tumor size, cm (N = 38) Median (range) 2.45 (0.3-7.0) Differentiation grade (N = 34) Well differentiated 3 (8.8) 1 (33.3) Moderately differentiated 24(70.6) 8(33.3) Poorly differentiated 7 (20.6) 2 (28.6) Microvascular invasion (N = 38) Present 19(50.0) 5(26.3) Absent 19(50.0) 6(31.6) Lymphatic invasion (N = 38) Present 21 (55.3) 5(23.8) Absent 17(44.7) 6(35.3) Perineural invasion (N = 38) Present 33(86.8) 11 (33.3) Absent 5(13.2) 0(0.0) Resection margin (N = 39) Present 1 (2.6) 1 (100.0) Absent 38(97.4) 10(26.3) PFS (univariable) HR (95% Cl) P value (N = 106) (event = 61) Total number Age Median (range) 1.01 (0.98-1.03) 0.712 Sex Female 1 Male 0.94(0.56-1.59) 0.823 Stage Resectable 1 (<0.001) Locally advanced 2.73(1.34-5.59) 0.006 Metastatic 4.23(2.19-8.17) <0.001 ECOG 0 1 1 + 2 1.24(0.74-2.06) 0.413 Tumor location Body and Tail 1 Head and Neck 0.70(0.42-1.16) 0.168 Initial treatment Surgery 1 (<0.001) CCRTorPBT 2.84(1.18-6.84) 0.020 Chemotherapy 3.13(1.47-6.67) 0.003 Supportive care only 9.25(4.13-20.71) <0.001 Operation name (N = 39) DP + TP 1 PD + PPPD 1.27(0.34-4.78) 0.728 Tumor size, cm (N = 38) Median (range) Differentiation grade (N = 34) Well differentiated 1 Moderately differentiated 1.43(0.18-11.49) 0.736 Poorly differentiated 1.30(0.12-14.37) 0.831 Microvascular invasion (N = 38) Present 1 Absent 0.91 (0.28-2.99) 0.873 Lymphatic invasion (N = 38) Present 1 Absent 1.35 (0.41-4.44) 0.617 Perineural invasion (N = 38) Present 1 Absent -- Resection margin (N = 39) Present 1 Absent 0.28 (0.04-2.33) 0.241 OS (univariable) Event (%) HR (95% Cl) (N = 106) Total number Age Median (range) 50(47.2) 1.01 (0.98-1.04) Sex Female 20(51.3) 1 Male 30(44.8) 0.87(0.49-1.54) Stage Resectable 6(14.6) 1 Locally advanced 17(68.0) 5.55(2.18-14.10) Metastatic 27(67.5) 8.94(3.64-21.97) ECOG 0 26(40.0) 1 1 + 2 24(58.5) 1.60(0.92-2.79) Tumor location Body and Tail 31 (58.5) 1 Head and Neck 19(35.9) 0.51 (0.29-0.91) Initial treatment Surgery 5(13.9) 1 CCRTorPBT 6(46.2) 2.75(0.84-9.01) Chemotherapy 20(60.6) 6.44(2.35-17.64) Supportive care only 19(79.2) 22.41 (7.99-62.81) Operation name (N = 39) DP + TP 2 (15.4) 1 PD + PPPD 3(11.5) 0.61 (0.10-3.66) Tumor size, cm (N = 38) Median (range) Differentiation grade (N = 34) Well differentiated 0(0.0) 1 Moderately differentiated 4(14.7) Poorly differentiated 1 (14.3) -- Microvascular invasion (N = 38) Present 3(15.8) 1 Absent 2 (10.5) 0.42 (0.07-2.60) Lymphatic invasion (N = 38) Present 3(14.3) 1 Absent 2(11.8) 0.54(0.09-3.38) Perineural invasion (N = 38) Present 5(15.2) 1 Absent 0(0.0) -- Resection margin (N = 39) Present 1 (100.0) 1 Absent 4(10.5) 0.08(0.01-0.95) OS (univariable) P value (event = 50) Total number Age Median (range) 0.503 Sex Female Male 0.641 Stage Resectable (<0.001) Locally advanced <0.001 Metastatic <0.001 ECOG 0 1 + 2 0.097 Tumor location Body and Tail Head and Neck 0.023 Initial treatment Surgery (<0.001) CCRTorPBT 0.096 Chemotherapy <0.001 Supportive care only <0.001 Operation name (N = 39) DP + TP PD + PPPD 0.588 Tumor size, cm (N = 38) Median (range) Differentiation grade (N = 34) Well differentiated Moderately differentiated Poorly differentiated Microvascular invasion (N = 38) Present Absent 0.351 Lymphatic invasion (N = 38) Present Absent 0.508 Perineural invasion (N = 38) Present Absent Resection margin (N = 39) Present Absent 0.045 ECOG, Eastern Cooperative Oncology Group; CCRT, computer- controlled radiation therapy; PBT, proton beam therapy; DP, distal pancreatectomy; TP, total pancreatectomy; PD, pancreaticoduodenectomy; PPPD, pylorus-preserving pancreaticoduodenectomy. Table 2. Univariable and multivariable analyses of OS and PFS for tumor biomarkers at baseline. Univariable model PFS Variables Cut point N Event (%) HR (95% Cl) CA19-9, U/mL [less than or 40 21 (52.5) 1 equal to] 37 >37 66 40 (60.6) 1.40 (0.82-2.38) CEA, U/mL [less than or 58 29 (50.0) 1 equal to]5 >5 48 32 (66.7) 1.70 (1.03-2.82) Cell-free DNA [less than or 53 30 (56.6) 1 concentration, equal to]427 ng/mL >427 53 31 (58.5) 0.88 (0.53-1.46) KRAS mutation [less than or 53 23 (43.4) 1 concentration, equal to] 0.165 copies/pL >0.165 53 38 (71.7) 2.64(1.55-4.50) KRAS mutation [less than or 53 24(45.3) 1 fractional equal to]0.415 abundance, % >0.415 53 37 (69.8) 2.18 (1.30-3.66) Univariable model PFS Variables Cut point P value Event (%) CA19-9, U/mL [less than or 15(37.5) equal to] 37 >37 0.214 35 (53.0) CEA, U/mL [less than or 23(39.7) equal to]5 >5 0.039 27 (56.3) Cell-free DNA [less than or 24(45.3) concentration, equal to]427 ng/mL >427 0.619 26(49.1) KRAS mutation [less than or 19(35.9) concentration, equal to] 0.165 copies/pL >0.165 <0.001 31 (58.5) KRAS mutation [less than or 19(35.9) fractional equal to]0.415 abundance, % >0.415 0.003 31 (58.5) Univariable model OS Variables Cut point HR (95% Cl) P value CA19-9, U/mL [less than or 1 equal to] 37 >37 2.03 (1.10-3.75) 0.023 CEA, U/mL [less than or 1 equal to]5 >5 1.63 (0.93-2.84) 0.087 Cell-free DNA [less than or 1 concentration, equal to]427 ng/mL >427 0.86 (0.49-1.50) 0.588 KRAS mutation [less than or 1 concentration, equal to] 0.165 copies/pL >0.165 2.54(1.41-4.56) 0.002 KRAS mutation [less than or 1 fractional equal to]0.415 abundance, % >0.415 2.19 (1.23-3.90) 0.008 Multivariable model (a) PFS Variables Cut point HR (95% Cl) P value CA19-9, U/mL [less than or 1 equal to] 37 >37 1.13 (0.64-1.98) 0.681 CEA, U/mL [less than or 1 equal to]5 >5 1.23 (0.70-2.16) 0.476 Cell-free DNA [less than or 1 concentration, equal to]427 ng/mL >427 1.03 (0.61-1.72) 0.923 KRAS mutation [less than or 1 concentration, equal to] 0.165 copies/pL >0.165 2.08 (1.20-3.63) 0.009 KRAS mutation [less than or 1 fractional equal to]0.415 abundance, % >0.415 1.73 (1.02-2.95) 0.042 Multivariable model (a) OS Variables Cut point HR (95% Cl) P value CA19-9, U/mL [less than or 1 equal to] 37 >37 1.51 (0.81-2.82) 0.195 CEA, U/mL [less than or 1 equal to]5 >5 0.93(0.51-1.70) 0.818 Cell-free DNA [less than or 1 concentration, equal to]427 ng/mL >427 1.07 (0.61-1.89) 0.804 KRAS mutation [less than or 1 concentration, equal to] 0.165 copies/pL >0.165 1.97 (1.05-3.67) 0.034 KRAS mutation [less than or 1 fractional equal to]0.415 abundance, % >0.415 1.73(0.95-3.14) 0.075 (a) Stage was adjusted in multivariable models of PFS and OS.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Molecular Diagnostics and Genetics|
|Author:||Kim, Min Kyeong; Woo, Sang Myung; Park, Boram; Yoon, Kyong-Ah; Kim, Yun-Hee; Joo, Jungnam; Lee, Woo|
|Article Type:||Clinical report|
|Date:||Apr 1, 2018|
|Previous Article:||Advanced Whole-Genome Sequencing and Analysis of Fetal Genomes from Amniotic Fluid.|
|Next Article:||Resolution of Spurious Immunonephelometric IgG Subclass Measurement Discrepancies by LC-MS/MS.|