Methods: A total of 221 plasma samples were collected from 84 HNSCC patients under informed consent and IRB approval. cfDNA was extracted from plasma, ~20M paired-end NGS mappable reads (reference: HG19) per sample were generated and CNI scores were calculated by read counting statistics. After unblinding CNI scores were evaluated as a diagnostic parameter for association with disease characteristics and progression. Survival analysis was conducted after dichotomization of the baseline CNI scores at a value of 27 corresponding to the 97.5th percentile of a normal reference group (RG, N = 142). Background: Copy number instability (CNI) signatures of cancers can be readily detected by Next Generation Sequencing of plasma cell-free DNA (cfDNA). HPV detected in oropharyngeal carcinomas is currently the only prognostic biomarker available. We report here CNI scores for disease monitoring of Head and Neck Squamous Cell Carcinomas (HNSSC) with potential predictive value for personalized therapeutic options. Results: CNI scores above the 97.5th RG percentile were detected in 41 out of 52 (79%) treatment naïve baseline samples. 27 patients with tumors ≤ T3 (73%, n=37) and 12 out of 14 (86%) with T4 tumors had CNI scores ≥ 27, with significantly higher CNI scores (p=0.027) seen for T4 tumors. Higher CNI scores were also found in patients with tumor lymph node invasion (n=36; median: 383) compared to negative lymph nodes (pN0, n=14; median: 28, p=0.002). High pre-operative CNIs (≥27) correlated with progression-free survival (PFS, Kaplan-Meier log-rank p=0.032). The median time to progression was 28 months for patients with a pre-operative CNI score ≥27 and was not reached for those with a score <27 (median follow-up 60m). In Cox regression, the pre-operative CNI scores were a stronger predictor of PFS (p=0.004) than any of the other analyzed clinical feature. A steep decline of CNI scores was detected after surgical resection, with increasing CNI scores in later disease progression. NORMAL REFERENCE THRESHOLD Parameter HR (95% CI) P value Age 1.004 (0.954-1.058) 0.867 Sex Female vs male 1.292 (0.414-4.029) 0.659 Smoking and/or alcohol Yes vs No 1.055 (0.300-3.176) 0.933 HNSCC history at site Yes vs No 0.979 (0.222-4.320) 0.978 T stage pT3/pT4 vs pT1/pT2 3.085 (0.980-9.714) 0.054 N stage pN2/pN3 vs pN0/pN1 1.985 (0.689-5.722) 0.204 CD vs no CD/pN0 1.023 (0.355-2.945) 0.967 Disease stage IV vs I/II/III 2.094 (0.674-6.500) 0.201 Tumor Grading G3 vs G1/G2 0.716 (0.228-2.250) 0.568 Adjuvant RCT as intended § Yes vs No 0.248 (0.087-0.704) 0.009 CNI score 1.591 (1.164-2.176) 0.004 Parameter HR (95% CI) P value pT3/pT4 vs pT1/pT2 1.544 (0.431-5.539) 0.505 Adjuvant RCT as intended 0.238 (0.073-0.783) 0.018 CNI score 1.563 (1.127-2.167) 0.007 Figure 3: ROC-curve analysis for CNI scores obtained from pre-surgery plasma cfDNA samples of 52 HNSCC patients versus 142 apparently tumor-free controls. Figure 4: Kaplan-Meier plots for progression-free survival stratified by pre-surgical CNI scores at a threshold of 27 (97.5th percentile of normal group), statistically assessed by log-rank test. Table 1 & 2: Univariate and multivariate Cox regression analyses for progression-free survival in fully assessable patients (N=52). Patient and disease characteristics and the pre- surgery CNI score (1000 units in log scale) were tested separately for relation with PFS. Parameters with P ≤ 0.05 in univariate analysis (left) were considered in a multivariate (right) Cox regression model. Higher CNI scores were superior to clinical scoring to predict shorter PFS with a hazard ratio of 1.56 (95% CI: 1.13-2.17) for CNI increase of 1000. CD = capsular disruption, RCT = radio-chemotherapy Figure 5: CNI scores in pre-surgery plasma samples stratified by A: Tumor size; the median CNI scores were 11 (n=142), 40 (n=5), 43 (n=15), 102 (n=17), 1049 (n=14) for controls w/o tumor, pT1, pT2, pT3, and pT4 tumors, B: pN status; 28 (n=14), 76 (n=19), and 1393 (n=17) for pN0, pN+ w/o, and pN+ with capsular spread, C: Disease stage: 25 (n=7), 28 (n=11), and 528 (n=33) for stage 1+2, 3, and 4, D: Grading: 60 (n=37) and 496 (n=15) for G1/G2 (only one with G1) and G3. Bold horizontal line: Median; Boxes: 25th/75th percentile; Whiskers: Minimum and maximum values. A D Figure 1: CONSORT diagram Figure 2: Method of CNI score determination Figure 6: CNI scores during treatment course for 67 eligible HNSCC patients. Median CNI scores were 99 (n=52), 23 (n=25), 18 (n=24), 21 (n=20), 20 (n=19), 21 (n=10), 23 (n=11), and 17 (n=5) for samples obtained pre-surgery, post-surgery, mid RT, end RT, 3-9, 10-18, 19-36, and > 36 months after surgery. Inset: direct comparison of patients with paired pre- surgery and post-surgery data (N=15). Figure 7: Time course of individual CNI scores and clinical events in two patients who developed metastatic disease and two patients without local recurrences or distant metastases. Figure 8: Investigation of tumor tissue samples from two patients with low pre-surgery CNI scores. Significant aberrations were detected in both tumors indicating the absence of detectable circulating tumor DNA rather than the absence of copy-number variation in the tumor genome. Conclusion: Chromosomal instability within HNSCC was quantified from plasma cfDNA as CNI score. The CNI score may serve as a better predictor for disease-specific outcome than common clinical staging parameters. The data suggests a promising biomarker for individual disease stratification and outcome with major implications on treatment decisions. B C Cell-free DNA for treatment monitoring and outcome predictor in head and neck cancer Julia Beck 1 , Markus Schirmer 2 , Martin Leu 2 , Margret Rave-Fraenk 2 , Howard Urnovitz 1 , Kirsten Bornemann-Kolatzki 1 , William M. Mitchell 3 , Michael Oellerich 4 , Ekkehard Schütz 1 , Martin Canis 5 1 Chronix Biomedical, Göttingen, Germany; 2 Clinic for Radiotherapy and Radiation Oncology, University Medical Center, Göttingen, Germany, 3 Department of Pathology, Vanderbilt University, Nashville, TN; 4 Institute of Clinical Pharmacology, University Medical Center, Göttingen, Germany; 5 Department of Otorhinolaryngology, Head and Neck Surgery, University Medical Center, Göttingen, Germany 100,00 0 CNI score Controls w/o tumor After surgery before RT End RT 3-9 m post OP CNI score Pre-surgery Mid RT 10-18 m post OP 19-36 m post OP > 36 m post OP 10,00 0 P=0.031 After surgery before RT Presurgery 10,00 0 1,00 0 1 10 10 0 1,00 0 100 10 1 RT = radiotherapy OP = operation/surgery Pa$ents (N=84) with suspected HNSCC and any CNI score assessed No HNSCC (N=1) HNSCC w/o second cancer (N=79) HNSCC (N=83) HNSCC with distant metastases (N=4) HNSCC w/o distant metastases (N=75) Simultaneous second cancer (N=4) Primary RCT (N=5) Eligible pa*ents: Surgery, adjuvant R(C)T (N=65) No CNI prior to surgery (N=13) Fully assessable pa*ents (N=52) Surgery incomplete/cancelled (N=4) Subjected to surgery (N=70) Sampling error (N=1)