Predictor Of Early Relapse Research Articles

Abstract 3658: Longitudinal tumor-informed ctDNA monitoring with the originally developed OTS-Assay system

Abstract Background: Cancer Genome Profiling (CGP) tests are now widely available in daily practice. However, only a very small fraction of patients have the opportunity for genome-guided therapies. Somatic mutations from cancer specimens are unique to individual patients and hold utility as individual tumor markers in blood. We have developed an original circulating tumor DNA (ctDNA) monitoring system, called OTS (Off-The-Shelf)-Assay, at Iwate Medical University Hospital, Yahaba, Japan. Patients and Methods: Between April 2022 and October 2023, 79 patients who received the OTS-Assay were enrolled in the present analysis. The OTS-Assay includes three steps: (a) OTS-Scan (somatic mutation identification by either tissue or blood), (b) OTS-Select (somatic mutation selection), and (c) OTS-Monitor (periodical somatic mutation quantification as ctDNA). For the OTS-Select, we established an originally developed >90 stepwise selection algorithm. One of the determinants is the availability of appropriate digital PCR (dPCR) probes for individual patients. Here, we search the probes for individual patients from the dPCR probe library, OTS-1000ex, which consists of >1000 probes against frequently reported human somatic mutations (Quantdetect, Tokyo, Japan). Results: All 79 patients were subject to the analysis. The cancer stage of all the patients was advanced. More than 10 tumor types were enrolled, including esophagus (n=42), colorectal (n=12), breast (n=7), lung (n=4), pancreas (n=3), hematological (n=3), and others (n=8). Fifty patients received OTS-Scan, and 29 patients had received a CGP test before the OTS-Assay (i.e., OTS-Scan was not needed). Seventy-five patients received OTS-Select, meaning that four patients had somatic mutations that were designated for ctDNA monitoring. All 79 patients received at least one OTS-monitor. The number of OTS-Monitors received ranged from 1 to 11 during the 19-month period, and the average interval was 3.0 months. The average pretreatment variant allele frequency (VAF) was 4.96% and the majority of the follow-up VAF was less than 1%. More than 90% of the patients had at least one clinical validity defined by: (a) early relapse prediction, (b) treatment efficacy evaluation, or (c) no relapse corroboration. Conclusion: The OTS-Assay system provides stable longitudinal ctDNA monitoring at the range of less than 1% of VAF using the originally developed dPCR probe library. The sensitivity and affordability of the OTS-Assay system allows frequent monitoring, which leads to obtaining important clinical information in the context of advanced cancer therapy. Citation Format: Satoshi S. Nishizuka, Hayato Hiraki, Akiko Yashima-Abo, Takeshi Iwaya. Longitudinal tumor-informed ctDNA monitoring with the originally developed OTS-Assay system [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3658.

Personalized circulating tumor DNA monitoring improves recurrence surveillance and management after curative resection of colorectal liver metastases: a prospective cohort study.

Approximately 60% of patients with colorectal liver metastases (CRLM) experience relapse within 2 years after radical resection, previous studies have proven that repeat local treatment (LT) could prolong survival, however, it is difficult to seize the window for LT due to the lack of a high-sensitive surveillance method. In this study, the authors aim to examine the value of longitudinal circulating tumor DNA (ctDNA) in guiding adjuvant chemotherapy, optimizing clinical surveillance strategy, and thereby improving CRLM outcomes. The authors conducted a prospective clinical trial using a personalized, tumor-informed ctDNA assay to monitor 60 CRLM patients undergoing resection with curative intent. Formalin-fixed paraffin-embedded tumor samples were collected after surgery. Blood samples were collected before surgery, 30 days after surgery (post-OP), and every third month until relapse or up to 2 years. A total of 394 plasma samples from 60 eligible patients were analyzed, with a median follow-up time of 31.3 months. Landmark analyses revealed that detectable ctDNA at post-OP (HR, 4.8), postadjuvant chemotherapy (HR, 6.0), and end-of-treatment (HR, 5.6) were associated with higher recurrence risk ( P <0.001). Post-OP ctDNA positivity served as the only independent prognostic marker in the multivariant analysis (HR, 5.1; P <0.001). Longitudinal ctDNA analysis identified relapsed patients at both sensitivity and specificity of 100%. Most (75%) patients were found with radiological relapse within 6 months after the first detectable ctDNA with a median lead time of 3.5 months. In relapsed patients, 73.2% had oligometastatic disease and 61% were liver-restricted, of which 72.0% received repeat LTs, and 60.0% achieved a secondary no evidence of disease status. Longitudinal ctDNA monitoring assists in early prediction of relapse, and thereby improves survival of CRLM patients by increased secondary resection rate and secondary no evidence of disease rate.

Utility of Next-Generation Sequencing-Based Chimerism Analysis for Early Relapse Prediction following Allogenic Hematopoietic Cell Transplantation.

Chimerism monitoring following allogeneic hematopoietic cell transplantation (HCT) plays a pivotal role in evaluating engraftment status and identifying early indicators of relapse. Recent advancements in next-generation sequencing (NGS) technology have introduced AlloSeq HCT as a more sensitive alternative to short tandem repeat (STR) analysis. This study aimed to compare AlloSeq HCT with STR, focusing on the prediction of early relapse post-allogeneic HCT. Chimerism levels in 29 HCT recipients were assessed using both STR and NGS, employing a total of 125 whole blood or bone marrow aspirate samples (68 post-HCT and 57 pre-HCT samples from recipients or donors). AlloSeq HCT exhibited high concordance with STR and demonstrated the potential for early detection of chimeric changes, particularly at extremely low levels. The combined advantages of high sensitivity and automated data analysis offered by AlloSeq HCT substantiate its clinical adoption for effective chimerism monitoring.

Circulating ESR1, long non-coding RNA HOTAIR and microRNA-130a gene expression as biomarkers for breast cancer stage and metastasis

Breast cancer, the most prevalent cancer among women, has posed a significant challenge in identifying biomarkers for early diagnosis and prognosis. This study aimed to elucidate the gene expression profile of Estrogen Receptor-1 (ESR-1), long non-coding RNA HOTAIR, and microRNA-130a in the serum of Egyptian breast cancer patients, evaluating the potential of HOTAIR and miR-130a as biomarkers for predicting pathological parameters in BC. The study involved 45 patients with primary BC, with serum samples collected preoperatively and postoperatively twice. The expression levels of ESR-1, HOTAIR, and miR-130a were quantified using real-time PCR and analyzed for correlations with each other and with the clinical and pathological parameters of the patients. Serum HOTAIR levels exhibited a strong positive association with metastasis and demonstrated a significant increase after 6 months in all patients with locally advanced and stage IV BC. Conversely, tumors with advanced stages and metastatic lesions showed significantly lower expression levels of miR-130a. Notably, a significant positive correlation was observed between preoperative ESR-1 expression and both HOTAIR and miR-130a levels. Serum HOTAIR and miR-130a levels have emerged as promising non-invasive biomarkers with the potential to predict the pathological features of BC patients. HOTAIR, an oncogenic long non-coding RNA (lncRNA), and miR-130a, a tumor suppressor miRNA, play crucial roles in tumor progression. Further investigations are warranted to elucidate the intricate interplay between HOTAIR and miR-130a and to fully comprehend the contribution of HOTAIR to BC recurrence and its potential utility in early relapse prediction.

Identification and surveillance of rare relapse-initiating stem cells during complete remission after transplantation

AbstractRelapse after complete remission (CR) remains the main cause of mortality after allogeneic stem cell transplantation for hematological malignancies and, therefore, improved biomarkers for early prediction of relapse remains a critical goal toward development and assessment of preemptive relapse treatment. Because the significance of cancer stem cells as a source of relapses remains unclear, we investigated whether mutational screening for persistence of rare cancer stem cells would enhance measurable residual disease (MRD) and early relapse prediction after transplantation. In a retrospective study of patients who relapsed and patients who achieved continuous-CR with myelodysplastic syndromes and related myeloid malignancies, combined flow cytometric cell sorting and mutational screening for persistence of rare relapse-initiating stem cells was performed in the bone marrow at multiple CR time points after transplantation. In 25 CR samples from 15 patients that later relapsed, only 9 samples were MRD-positive in mononuclear cells (MNCs) whereas flowcytometric-sorted hematopoietic stem and progenitor cells (HSPCs) were MRD-positive in all samples, and always with a higher variant allele frequency than in MNCs (mean, 97-fold). MRD-positivity in HSPCs preceded MNCs in multiple sequential samples, in some cases preceding relapse by >2 years. In contrast, in 13 patients in long-term continuous-CR, HSPCs remained MRD-negative. Enhanced MRD sensitivity was also observed in total CD34+ cells, but HSPCs were always more clonally involved (mean, 8-fold). In conclusion, identification of relapse-initiating cancer stem cells and mutational MRD screening for their persistence consistently enhances MRD sensitivity and earlier prediction of relapse after allogeneic stem cell transplantation.

17th International Conference on Malignant Lymphoma, Palazzo dei Congressi, Lugano, Switzerland, 13 - 17 June, 2023.

Introduction: Circulating tumor DNA (ctDNA) has been proven to be a promising tumor-specific biomarker in tumors, but its clinical utility in risk stratification and early prediction of relapse for peripheral T cell lymphoma (PTCL) has not been well explored. Methods: Using a lymphoma-specific sequencing panel, we assessed the prognostic and predictive utilities of ctDNA measurements before and after first-line therapy in 41 PTCL patients. Results: For the 41 patients, we obtained a total of 36 primary tumor specimens, 41 pretreatment blood samples and 21 serial blood samples before and after first-line therapy. By the last visit in March 2023, the median follow-up duration was 10.0 (range, 2.5–18.8) months. The clinical characteristics of the 41 patients were shown in Figure 1A. All patients received CHOP-like regimen. The top 30 somatic mutations of tissues were shown in Figure 1B. Pretreatment ctDNA (Pre-ctDNA) achieved an overall sensitivity of 81.3% (159/196) in detecting variants verified in tumor, indicating that ctDNA is a reliable source for PTCL genotyping. In addition, ctDNA allowed for the identification of additional 98 somatic mutations that were undetectable in tumor gDNA, which demonstrated that ctDNA could overcome tumor spatial heterogeneity (Figure 1C). Pre-ctDNA burden was significantly associated with clinical characteristics, including extranodal involvement (P < 0.001), lactate dehydrogenase (LDH) levels (P < 0.001), age (P = 0.047), B symptoms (P = 0.038), stage (P = 0.008) and international prognosis index (IPI) score (P < 0.001) (Figure 1D). The cell-free (cfDNA) fragments profoundly reflects both genomic and chromatin characteristics. Our results suggested that all samples had a more prominent mono-nucleosomal fragments abundance (167 bp), whereas the samples of angioimmunoblastic T-cell lymphom (AITL) and the samples with high IPI scores or non-complete response (CR) in the mid-stage of treatment had a more prominent shift towards shorter cfDNA size (Figure 1E). Similarly, when we analyzed the response to treatment, significant decline in ctDNA levels was observed at end of treatment (EOT) in patients with CR compared with those who did not achieved CR. The 4 patients with CR and EOT ctDNA negative achieved continuous CR until now (Figure 1F). Furthermore, high pre-ctDNA levels presented unfavourable PFS (P = 0.002) and OS (P = 0.020) (Figure 1G). Multivariate cox regression analysis showed that high pre-ctDNA burden (P = 0.029) and TP53 mutation (P = 0.005) (Figure 1H). Subgroup analysis showed that patients without these 2 risk factors (Type 1) had longer PFS than the other patients with more than 0 risk factor (Type 2) (P < 0.001) (Figure 1I). Keywords: Aggressive T-cell non-Hodgkin lymphoma, Diagnostic and Prognostic Biomarkers No conflicts of interests pertinent to the abstract.

Postoperative Relapse Prediction in Patients With Ewing Sarcoma Using Computed Tomography-Based Radiomics Models Covering Tumor Per Se and Peritumoral Signatures.

We aimed to develop and validate a computed tomography (CT)-based radiomics model for early relapse prediction in patients with Ewing sarcoma (ES). We recruited 104 patients in this study. Tumor areas and areas with a tumor expansion of 3 mm were used as regions of interest for radiomics analysis. Six different models were constructed: Pre-CT, CT enhancement (CTE), Pre-CT +3 mm , CTE +3 mm , Pre-CT and CTE combined (ComB), and Pre-CT +3 mm and CTE +3 mm combined (ComB +3 mm ). All 3 classifiers used a grid search with 5-fold cross-validation to identify their optimal parameters, followed by repeat 5-fold cross-validation to evaluate the model performance based on these parameters. The average performance of the 5-fold cross-validation and the best one-fold performance of each model were evaluated. The AUC (area under the receiver operating characteristic curve) and accuracy were calculated to evaluate the models. The 6 radiomics models performed well in predicting relapse in patients with ES using the 3 classifiers; the ComB and ComB +3 mm models performed better than the other models (AUC -best : 0.820-0.922/0.823-0.833 and 0.799-0.873/0.759-0.880 in the training and validation cohorts, respectively). Although the Pre-CT +3 mm , CTE +3 mm, and ComB +3 mm models covering tumor per se and peritumoral CT features preoperatively forecasted ES relapse, the model was not significantly improved. The radiomics model performed well for early recurrence prediction in patients with ES, and the ComB and ComB +3 mm models may be superior to the other models.

A novel risk model for predicting early relapse in acute myeloid leukemia patients undergoing allogeneic hematopoietic stem-cell transplantation.

Relapse remains the leading cause of death in acute myeloid leukemia (AML) patients following allogeneic hematopoietic stem-cell transplantation (allo-HSCT), limiting the efficacy of allo-HSCT. Thus, the ability to identify high-risk patients in a manner that permits early intervention has the potential to improve survival outcomes. We retrospectively enrolled 414 younger patients (aged 14-60 years) with AML who received allo-HSCT between January 2014 and May 2020. From June 2020 to June 2021, 110 consecutive patients were included prospectively in the validation cohort. The primary outcome was early relapse (relapse within 1 year). The cumulative incidence of early relapse after allo-HSCT was 11.8%. The overall survival rate for patients who relapsed within 1-year was 4.1% at 3 years after relapse. After multivariable adjustment, statistically significant associations between primary resistance, pre-transplantation measurable residual disease, DNMT3A mutation, or white blood cell count at diagnosis and early relapse were observed. An early relapse prediction model was developed based on these factors and the model performed well. Patients deemed to have a high risk or a low risk of early relapse had early relapse rates of 26.2% and 6.8%, respectively (P < 0.001). The prediction model could be used to help identify patients at risk for early relapse and to guide personalized relapse prevention.

Abstract 2208: Rare neoplastic subpopulation predicts relapse in nasopharyngeal carcinoma

Abstract Nasopharyngeal carcinoma (NPC), with the highest incidence and prevalence in Southern China including Hong Kong, originates from the nasopharyngeal epithelium and frequently metastasises to cervical lymph nodes and distal organs such as liver. Radiation therapy with/without concurrent chemoradiation is the most common first-line treatment. Although most patients respond to the treatment, about 30-50% of treated patients still relapse. Bulk sequencing uncovered numerous disease-associated genes, but still there is no effective targeted therapy or immunotherapy to prevent relapse. Tools for early relapse prediction at initial diagnosis still remain unavailable. Recently, single-cell RNA sequencing (scRNA-seq) technology has emerged as a sensitive long-sought solution to low resolution in bulk seqeuncing. We launched the largest and most comprehensive scRNA-seq NPC study in the world. 55 patients were recruited. Matched biopsies from normal adjacent tumour, primary tumour and lymph node malignancy sites were harvested and 70488 epithelial cells had been successfully scRNA-seq profiled. Unsupervised clustering identified 9 different epithelial subclusters. Copy number variation (CNV) analysis identified neoplastic cells with wide-spread CNV aberrations consistent with previous bulk genomic findings. For the first time, we had also attempted to unfold single nucleotide variation (SNV) of neoplastic cells from scRNA-seq using modified Mutect2 pipeline. Our study spanned 4 years and 5 patients were reported relapsed. Encouragingly, survival analysis identified rare neoplastic subcluster 1 as the only neoplastic subcluster positively correlated with relapse. Abundance of neoplastic subcluster 1 biomarker-positive cells were verified to be elevated in FFPE primary tumours of relapsed patients collected at initial diagnosis. Deconvolution analysis of a publicly available bulk RNA sequencing dataset revealed only cells harbouring neoplastic subcluster 1 expression signature tended to positively correlate with relapse. Interestingly, cross-tissue abundance analysis identified neoplastic subcluster 1 was positively correlated with preferential colonialization at cervical lymph node. Single-cell gene-set enrichment analysis (scGSEA) suggested neoplastic subcluster 1 possessed higher metastasis and treatment resistance potential. For the first time, we had also investigated the cell-cell communications among epithelial cells and identified interaction pathways implicated in self-renewal and tumorigenic capacity of cancer-initiating cells to be associated with neoplastic subcluster 1 and relapse. To conclude, we successfully discovered a previously overlooked rare neoplastic subpopulation in NPC that may (1) possess higher metastasis and treatment resistance capacity, and (2) play a regulatory role in tumourigenesis, and its abundance at initial diagnosis could predict relapse. Citation Format: Ka Chun Wu, Ken Yu, Nelson Su, Jeffrey Yan Ho Lau, Anthony Lo, Yvonne Ngar Woon Kam, Raymond Tsang, Joseph Chung, Joshua Ho, Chi Ming Che, Victor Ho Fun Lee. Rare neoplastic subpopulation predicts relapse in nasopharyngeal carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2208.

Abstract 3326: The clinical validity of urinary pellet DNA monitoring of recurrent bladder cancer

Abstract Introduction: Approximately half of high-risk non-muscle-invasive bladder cancers (NMIBCs) recur after transurethral resection of the bladder tumor (TURBT), whereas conventional urological testing for bladder cancer (BC) patients has been considered invasive and with low sensitivity. Our previous reports have shown that circulating tumor DNA (ctDNA) monitored by digital PCR (dPCR) can predict relapse and evaluate treatment efficacy in digestive tract cancers. In BC patients, tumor-derived genetic mutations detected in urinary DNA are expected to be a diagnostic biomarker. This study evaluates the clinical validity of monitoring tumor-specific gene mutations in ctDNA and urine pellet DNA (upDNA) using dPCR as a biomarker for early relapse prediction and determination of treatment efficacy. Materials and Methods: Thirty-two previously treated and untreated BC patients were enrolled. Tumor DNA was extracted from archived TURBT tissues, and gene mutation analysis by Next Generation Sequencing (NGS) of tumors was performed to select patient-unique somatic mutations for both ctDNA and upDNA monitoring by dPCR. Longitudinal variant allele frequencies (VAFs) of ctDNA and upDNA were monitored by dPCR for up to two years. While dPCR is rapid and has a high sensitivity of less than 0.1% VAF, a mutation-specific primer/probe set is required. Therefore, we established over 1,000 primer/probe sets for frequent mutations in human cancer. This large-scale dPCR primer/probe library allow us to select patient specific mutations for immediate ctDNA monitoring. Clinical recurrence (Crec) based on imaging and urinary cell findings are compared with VAF dynamics of ctDNA and upDNA. Results: The median observation period was 516 days (30-733), and a total of 230 urine samples were collected. The selection of monitoring mutations based on NGS analysis was performed in 93.8% (30/32) cases, with 2.3 (range: 1-6) mutations monitored per case. The primer/probe sets covered 87.5% (28/32) of cases and 87.5% (42/48) of mutations. The VAFs of the upDNA of cases started showing constantly increasing trends 3-12 months earlier than the diagnosis of Crec in 71.4% (5/7) of the cases. The other two Crec cases, which did not show an elevated upVAF, were both pyuria cases. After local treatment, the upDNA VAFs remained high in the Crec cases. Conventional Crec did not seem to be fully reflected by ctDNA alone. Conclusion: Frequent VAF monitoring using upDNA by dPCR enables the prediction of local relapse and the evaluation of treatment efficacy in the management of BC patients. Citation Format: Masakazu Abe, Hayato Hiraki, Takashi Tsuyukubo, Sadahide Ono, Shigekatsu Maekawa, Daichi Tamura, Daiki Ikarashi, Renpei Kato, Tomohiko Matsuura, Mitsugu Kanehira, Ryo Takata, Hiromitsu Fujisawa, Takeshi Iwaya, Masashi Idogawa, Wataru Obara, Satoshi S. Nishizuka. The clinical validity of urinary pellet DNA monitoring of recurrent bladder cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3326.

Long non-coding RNAs enable precise diagnosis and prediction of early relapse after nephrectomy in patients with renal cell carcinoma

PurposeRenal cell carcinoma belongs among the deadliest malignancies despite great progress in therapy and accessibility of primary care. One of the main unmet medical needs remains the possibility of early diagnosis before the tumor dissemination and prediction of early relapse and disease progression after a successful nephrectomy. In our study, we aimed to identify novel diagnostic and prognostic biomarkers using next-generation sequencing on a novel cohort of RCC patients.MethodsGlobal expression profiles have been obtained using next-generation sequencing of paired tumor and non-tumor tissue of 48 RCC patients. Twenty candidate lncRNA have been selected for further validation on an independent cohort of paired tumor and non-tumor tissue of 198 RCC patients.ResultsSequencing data analysis showed significant dysregulation of more than 2800 lncRNAs. Out of 20 candidate lncRNAs selected for validation, we confirmed that 14 of them are statistically significantly dysregulated. In order to yield better discriminatory results, we combined several best performing lncRNAs into diagnostic and prognostic models. A diagnostic model consisting of AZGP1P1, CDKN2B-AS1, COL18A1, and RMST achieved AUC 0.9808, sensitivity 95.96%, and specificity 90.4%. The model for prediction of early relapse after nephrectomy consists of COLCA1, RMST, SNHG3, and ZNF667-AS1 and achieved AUC 0.9241 with sensitivity 93.75% and specificity 71.07%. Notably, no combination has outperformed COLCA1 alone. Lastly, a model for stage consists of ZNF667-AS1, PVT1, RMST, LINC00955, and TCL6 and achieves AUC 0.812, sensitivity 85.71%, and specificity 69.41%.ConclusionIn our work, we identified several lncRNAs as potential biomarkers and developed models for diagnosis and prognostication in relation to stage and early relapse after nephrectomy.

Preoperative Mutational Analysis of Circulating Tumor Cells (CTCs) and Plasma-cfDNA Provides Complementary Information for Early Prediction of Relapse: A Pilot Study in Early-Stage Non-Small Cell Lung Cancer

We assessed whether preoperativemutational analyses of circulating tumor cells (CTCs) and plasma-cfDNA could be used as minimally invasive biomarkers and as complimentary tools for early prediction of relapse in early-stage non-small -cell lung cancer (NSCLC). Using ddPCR assays, hotspot mutations of BRAF, KRAS, EGFR and PIK3CA were identified in plasma-cfDNA samples and size-based enriched CTCs isolated from the same blood samples of 49 early-stage NSCLC patients before surgery and in a control group of healthy blood donors (n= 22). Direct concordance of the mutational spectrum was further evaluated in 27 patient-matched plasma-cfDNA and CTC-derived DNA in comparison to tissue-derived DNA. The prevalence of detectable mutations of the four tested genes was higher in CTC-derived DNA than in the corresponding plasma-cfDNA (38.8% and 24.5%, respectively).The most commonly mutated gene was PIK3CA, in both CTCs and plasma-cfDNA at baseline and at the time of relapse. Direct comparison of the mutation status of selected drug-responsive genes in CTC-derived DNA, corresponding plasma-cfDNA and paired primary FFPE tissues clearly showed the impact of heterogeneity both within a sample type, as well as between different sample components. The incidence of relapse was higher when at least one mutation was detected in CTC-derived DNA or plasma-cfDNA compared with patients in whom no mutation was detected (p =0.023). Univariate analysis showed a significantly higher risk of progression (HR: 2.716; 95% CI, 1.030-7.165; p =0.043) in patients with detectable mutations in plasma-cfDNA compared with patients with undetectable mutations, whereas the hazard ratio was higher when at least one mutation was detected in CTC-derived DNA or plasma-cfDNA (HR: 3.375; 95% CI, 1.098-10.375; p =0.034). Simultaneous mutational analyses of plasma-cfDNA and CTC-derived DNA provided complementary molecular information from the same blood sample and greater diversity in genomic information for cancer treatment and prognosis. The detection of specific mutations in ctDNA and CTCs in patients with early-stage NSCLC before surgery was independently associated with disease recurrence, which represents an important stratification factor for future trials.

Impact of Ultra High-risk Genetics on Real-world Outcomes of Transplant-eligible Multiple Myeloma Patients.

Refined prediction of early relapse following standard-of-care (SoC) autologous stem cell transplant (ASCT) in newly diagnosed multiple myeloma (NDMM) could inform real-world risk-stratified post-ASCT strategies. We investigated the impact of double hit genetics (≥2 adverse markers: t(4;14), t(14;16), t(14;20), gain(1q), del(17p)) on outcome in 139 NDMM patients who underwent SoC ASCT between January 2014 and October 2019 at our center. Double hit genetics were associated with a significantly shortened progression-free survival (hazard ratio [HR] = 4.27, P < 0.001) and overall survival (HR = 4.01, P = 0.03), and characterized most early relapses. Our results support the real-world utility of extended genetic profiling for improved risk prediction in NDMM.

Clinical implications of circulating tumor DNA in predicting the outcome of diffuse large B cell lymphoma patients receiving first-line therapy

BackgroundCirculating tumor DNA (ctDNA) has been proven to be a promising tumor-specific biomarker in solid tumors, but its clinical utility in risk stratification and early prediction of relapse for diffuse large B cell lymphoma (DLBCL) has not been well explored.MethodsHere, using a lymphoma-specific sequencing panel, we assessed the prognostic and predictive utilities of ctDNA measurements before, during, and after first-line therapy in 73 Chinese DLBCL patients.ResultsThe pretreatment ctDNA level serving as an independent prognostic factor for both progression-free survival (PFS, adjusted HR 2.47; p = 0.004) and overall survival (OS, adjusted HR 2.49; p = 0.011) was confirmed in our cohort. Furthermore, the patients classified as molecular responders who presented a larger decrease in ctDNA levels after the initial two treatment cycles had more favorable PFS (unreached vs. 6.25 months; HR 5.348; p = 0.0015) and OS (unreached vs. 25.87; HR 4.0; p = 0.028) than non-responders. In addition, interim ctDNA clearance may be an alternative noninvasive method of positron emission tomography and computed tomography (PET-CT) for predicting better PFS (HR 3.65; p = 0.0033) and OS (HR 3.536; p = 0.016). We also demonstrated that posttreatment ctDNA was a sensitive indicator for detecting minimal residual disease (MRD) in patients with a high risk of recurrence (HR 6.471; p = 0.014), who were otherwise claimed to achieve radiographic CR (complete remission).ConclusionsCtDNA is a promising noninvasive tool for prognosis prediction, response assessment, and early relapse prediction of first-line treatment in DLBCL patients.

Serum IgG N-glycans enable early detection and early relapse prediction of colorectal cancer.

Colorectal cancer (CRC) develops mainly from colorectal advanced adenomas (AA), which are considered precancerous lesions. Novel early diagnostic biomarkers are urgently needed to distinguish CRC and AA from healthy control (HC). Alternative glycosylation of serum IgG has been shown to be closely associated with CRC. We aimed to explore the potential of IgG N-glycan as biomarkers in the early differential diagnosis of CRC. The study population was strictly matched to the exclusion criteria process. Serum IgG N-glycan profiles were analyzed by a robust and reliable relative quantitative method based on ultra-performance liquid chromatography (UPLC). Relative quantification and classification performance of IgG N-glycans were evaluated by Mann-Whitney U tests and ROC curve based on directly detected and derived glycan traits, respectively. Six and 14 directly detected glycan traits were significantly changed in AA and CRC, respectively, compared with HC. GP1 and GP3 were able to accurately distinguish AA from HC for early precancerous lesions screening. GP4 and GP14 provided a high value in discriminating CRC from HC. A novel combined index named GlycoF, including GP1, GP3, GP4, GP14 and CEA was developed to provide a potential early diagnostic biomarker in discriminating simultaneously AA (AUC=0.847) and CRC (AUC=0.844) from HC. GlycoF also demonstrated a superior CRC detection rate across CRC all stages and conspicuous prediction ability of risk of relapse. Serum IgG N-glycans analysis provided powerful early screening biomarkers that can efficiently differentiate CRC and AA from HC.

340P Prediction of relapse in colon cancer patients by machine learning models combining radiomics and deep features extracted from baseline computed tomography

Up to 40% of localized colon cancer (CC) patients relapse despite optimal initial treatment. Circulating tumor DNA has emerged as a new prognostic biomarker in this setting. However, liquid biopsy tests provide a limited accuracy in the early prediction of relapse. On this basis, radiomics and artificial intelligence (AI) are posed to provide relevant insights in the prediction of relapse and patient outcome. Here we propose a predictive model of CC recurrence using clinical data in combination with radiomics and deep features extracted from baseline computed tomography (CT) images. A single-center retrospective observational study was designed. Real-world clinical data and CT examinations were collected from 2015 to 2017. Manual segmentation of the tumor was performed slice by slice by a radiologist with 10+ years of experience in CC using ITK-SNAP. Feature extraction techniques were applied to the tumor voxels through an analysis pipeline in Python adapted from Quibim Precision® software (Quibim, Valencia, ES). Feature reduction techniques were applied to select the characteristics providing independent information. The types of features extracted were: Radiomics, Deep features and fractal dimension. Several classifiers were trained with the clinical and image features as the input, to predict relapse from CC. Relapse and non-relapse classes were balanced to the number of slices containing tumor. Train-validation data ratio was 70:30. Baseline CT exams from 60 localized CC patients were included. 48% and 20% of them had stage III and high-risk stage II, respectively. The remaining patients were diagnosed with low-risk CC. 36.7% of patients relapsed. Merging the clinical data with the radiomics, fractal and deep features provided a high and increased accuracy 95% (CI 95%: 85 – 100) when compared to clinical variables alone (55%, CI 95%: 32 – 78). Random Forest and cross validation model provided the best performance. The extraction of radiomics and deep features from CT exams in CC patients and their combination in an AI model provide promising insights towards the early prediction of relapse, leading to identify new and accurate prognostic imaging panels.

Histological Features and Predictors of Early Relapse in Patients of Ulcerative Colitis

Objective: To determine the proportion of histological features and histological feature predicting frequency of elapse in known patients of ulcerative colitis. Methodology: One hundred and ninety cases each had endoscopic colonic biopsies from the rectum, distal colon, and proximal colon. Study Design: Retrospective Case series Setting: Sheikh Zayed Hospital Lahore, Department of Histopathology, from February 25, 2016, to February 25, 2017. Results: The majority of patients complained of generalized abdominal pain (55.3%). The commonest histological feature is cryptitis 189(99.5%). Majority patients with basal plasmacytosis 38(42.2%) and crypt abscess 36(29.8%) relapsed at 4th month. Conclusion: Histological features basal plasmacytosis and crypt abscess are of prognostic importance. A greater chance of relapse can be predicted in patients with increased number of basal plasma cells and cryptitis. These findings will help in further management of ulcerative colitis patients. Keywords: Inflammatory bowel disease, Ulcerative colitis, Remission, Relapse

Navigating preemptive and therapeutic donor lymphocyte infusions in advanced myeloid malignancies by high-sensitivity chimerism analysis.

Preemptive and therapeutic donor lymphocyte infusions (preDLI and tDLI) are widely used in relapsing and relapsed hematopoietic malignancies after allogeneic stem cell transplantation (alloSCT) to enhance the graft-versus-malignancy effect. However, in advanced myeloid malignancies, long-term survival after preDLI and tDLI remains low, reflecting our inability to master the double-edged sword of alloreactivity, balancing anti-neoplastic activity versus graft-versus-host disease (GvHD). We previously evaluated a quantitative PCR-based high-sensitivity chimerism (hs-chimerism) based on insertion/deletion polymorphisms instead of short tandem repeats, where increasing host chimerism in peripheral blood predicts relapse more than a month before clinical diagnosis, and declining host chimerism signals anti-host alloreactivity. Here we report 32 consecutive patients with advanced myeloid malignancies receiving preDLI or tDLI “navigated” by hs-chimerism (“navigated DLI”). We compared them to a historical cohort of 110 consecutive preDLI or tDLI recipients, prior to implementation of hs-chimerism at our institution (“controls”). Both groups were comparable regarding age, gender, conditioning, donor type, and time to DLI. With longer median follow-up of the navigated DLI group (8.5 versus 5 months), their landmark overall (64%) and disease-free survival (62%) at 2 years from first DLI compared favorably with controls (23% and 21%, respectively). Improved survival of navigated DLI was due to both reduced relapse incidence (38% versus 60%) and non-relapse mortality (17% versus 44%) at 2 years. Early relapse prediction by hs-chimerism allowed a preemptive approach in 28% of navigated DLI versus 7% in controls. Our results confirm hs-chimerism as a highly valuable tool for monitoring and steering immune interventions after alloSCT.

Longitudinal relapse pattern of patients with first-episode schizophrenia-spectrum disorders and its predictors and outcomes: A 10-year follow-up study.

This study explored the 10-year pattern of relapse of patients with first-episode schizophrenia-spectrum disorders (FES), predictors and outcomes of early and late relapse. Patients received EIS (N=148) in Hong Kong between 1 July 2001 and 30 June 2002 were matched with those who entered standard care (N=148) one year before. Relapse information over 10 years were obtained and 209 patients were interviewed at 10-year follow-up. Predictors of early relapse ([ER] relapse in initial three years) and late relapse group ([LR] relapse in year 4-10) and their differential outcomes were explored. Fifty-six patients (26.8%) were relapse-free over 10 years with more EIS patients. Among the relapsed patients, 63.6% were ER patients who had the poorest longitudinal outcomes, including higher suicide attempts, violence episodes, more hospitalization and lower employment, whereas the LR patients do not differ much from the no relapse group. Relapse-free patients required less hospitalization in the first episode and lower antipsychotic dosage. The LR patients had less positive symptoms in year one but longer first-episode hospitalization and higher antipsychotic dosage. Delaying the first relapse may help to improve the long-term outcomes. Good response to antipsychotic medications was associated with relapse-free over long-term. However, sufficient antipsychotic medications with good symptomatic control during the early stage of the illness is crucial for relapse prevention for other patients. These findings highlight illness heterogeneity and the importance in differential use of antipsychotics in relapse prevention.

Analysis of Subset Chimerism for MRD-Detection and Pre-Emptive Treatment in AML.

Allogeneic hematopoietic stem cell transplantation (alloHCT) represents the only potentially curative treatment in high-risk AML patients, but up to 40% of patients suffer from relapse after alloHCT. Treatment of overt relapse poses a major therapeutic challenge and long-term disease control is achieved only in a minority of patients. In order to avoid post-allograft relapse, maintenance as well as pre-emptive therapy strategies based on MRD-detection have been used. A prerequisite for the implementation of pre-emptive therapy is the accurate identification of patients at risk for imminent relapse. Detection of measurable residual disease (MRD) represents an effective tool for early relapse prediction in the post-transplant setting. However, using established MRD methods such as multicolor flow cytometry or quantitative PCR, sensitive MRD monitoring is only applicable in about half of the patients with AML and advanced MDS undergoing alloHCT. Donor chimerism analysis, in particular when performed on enriched leukemic stem and progenitor cells, e.g. CD34+ cells, is a sensitive method and has emerged as an alternative option in the post alloHCT setting. In this review, we will focus on the current strategies for lineage specific chimerism analysis, results of pre-emptive treatment using this technology as well as future developments in this field.