Sloan-Kettering Research Articles

Validation and Comparison of Prognostic Models in Renal Carcinoma in a Tertiary Hospital.

Renal cell carcinoma (RCC) is the third most frequent urological neoplasia. Proper risk stratification is essential for adequate management. Various calculators are available. This project aims to evaluate the accuracy of the calculators applied to our patients. We performed a retrospective study of the nephrectomies due to RCC performed from January 2008 to December 2013. We applied the most widely used predictive models (University of California, Los Angeles Integrated Staging System (UISS), Stage, Size, Grade and Necrosis (SSIGN), Memorial Sloan Kettering Cancer Center (MSKCC) and International Metastatic RCC Database Consortium (IMDC)) to stratify patients in different risk groups. We evaluated progression-free survival (PFS) or death caused by RCC (cancer-specific survival (CSS)) or other causes (overall survival (OS)). We analysed 238 patients. The 5-year OS, CSS and PFS were 76%, 85% and 83%, whereas the 10-year OS, CSS and PFS were 47%, 75% and 77%, respectively. The 5-year survival analysis by risk groups according to the prognostic models showed that the PFS was 0% and 20.4% in high- and intermediate-risk metastatic RCC (mRCC). Moreover, the PFS was 90%, 95.2% and 98.9% in localised high-, intermediate- and low-risk RCC according to the UISS (area under the receiver operating characteristics curve (AUC): 0.93). The SSIGN model showed a CSS of 99% for the group with the lowest score and 5.3% for the group with the worst prognosis (AUC: 0.91). The OS of mRCC showed medians of 13.25 and 87 months according to MSKCC (AUC: 0.75) and 16, 23 and 85 months according to IMDC (AUC: 0.71) (high risk, intermediate and low). The validation of the predictive models carried out with our patients showed consistency in many of the results. Risk stratification should be implemented.

Weakly supervised large-scale pancreatic cancer detection using multi-instance learning.

Early detection of pancreatic cancer continues to be a challenge due to the difficulty in accurately identifying specific signs or symptoms that might correlate with the onset of pancreatic cancer. Unlike breast or colon or prostate cancer where screening tests are often useful in identifying cancerous development, there are no tests to diagnose pancreatic cancers. As a result, most pancreatic cancers are diagnosed at an advanced stage, where treatment options, whether systemic therapy, radiation, or surgical interventions, offer limited efficacy. A two-stage weakly supervised deep learning-based model has been proposed to identify pancreatic tumors using computed tomography (CT) images from Henry Ford Health (HFH) and publicly available Memorial Sloan Kettering Cancer Center (MSKCC) data sets. In the first stage, the nnU-Net supervised segmentation model was used to crop an area in the location of the pancreas, which was trained on the MSKCC repository of 281 patient image sets with established pancreatic tumors. In the second stage, a multi-instance learning-based weakly supervised classification model was applied on the cropped pancreas region to segregate pancreatic tumors from normal appearing pancreas. The model was trained, tested, and validated on images obtained from an HFH repository with 463 cases and 2,882 controls. The proposed deep learning model, the two-stage architecture, offers an accuracy of 0.907 0.01, sensitivity of 0.905 0.01, specificity of 0.908 0.02, and AUC (ROC) 0.903 0.01. The two-stage framework can automatically differentiate pancreatic tumor from non-tumor pancreas with improved accuracy on the HFH dataset. The proposed two-stage deep learning architecture shows significantly enhanced performance for predicting the presence of a tumor in the pancreas using CT images compared with other reported studies in the literature.

Predicting prognosis outcomes of primary central nervous system lymphoma with high-dose methotrexate-based chemotherapeutic treatment using lipidomics.

Primary central nervous system lymphoma (PCNSL) is a rare extranodal lymphomatous malignancy which is commonly treated with high-dose methotrexate (HD-MTX)-based chemotherapy. However, the prognosis outcome of HD-MTX-based treatment cannot be accurately predicted using the current prognostic scoring systems, such as the Memorial Sloan-Kettering Cancer Center (MSKCC) score. We studied 2 cohorts of patients with PCNSL and applied lipidomic analysis to their cerebrospinal fluid (CSF) samples. After removing the batch effects and features engineering, we applied and compared several classic machine-learning models based on lipidomic data of CSF to predict the relapse of PCNSL in patients who were treated with HD-MTX-based chemotherapy. We managed to remove the batch effects and get the optimum features of each model. Finally, we found that Cox regression had the best prediction performance (AUC = 0.711) on prognosis outcomes. We developed a Cox regression model based on lipidomic data, which could effectively predict PCNSL patient prognosis before the HD-MTX-based chemotherapy treatments.

The Study of PIK3CA Hotspot Mutations and Co-Occurring with EGFR, KRAS, and TP53 Mutations in Non-Small Cell Lung Cancer.

PIK3CA-mutant non-small-cell lung cancer (NSCLC) is associated with other genetic mutations and may influence treatment strategies and clinical outcomes. We aimed to characterize PIK3CA mutations co-occurring with several major driver mutations using data from published cohorts and our medical center. We analyzed NSCLC patients harboring PIK3CA mutations from The Cancer Genome Atlas (TCGA) and Memorial Sloan Kettering (MSK) databases and retrospectively identified NSCLC patients with PIK3CA-mutants at a single medical center from our electronic records. The Log rank test was used to determine the association between PIK3CA mutations and overall survival (OS) in NSCLC patients. Common hotspot mutations in PIK3CA were found in exon 9 (c.1633G > A, E545K, and c.1624G > A, E542K) and exon 20 (c.3140A > G, H1047R) in all cohorts. Co-occurring mutations of PIK3CA with EGFR, KRAS, and TP53 have been frequently observed in patients with NSCLC, with different percentages in these datasets generated by different background. PIK3CA mutations were observed to be significantly associated with poor OS in lung adenocarcinomas patients in the MSKCC cohort (hazard ratio [HR] = 0.519, 95% confidence interval [CI] = 0.301-0.896; P <0.05). PIK3CA co-occurring mutations in other genes may represent distinct subsets of NSCLC. Further elucidation of the roles of PIK3CA hotspot mutations combined with other driver mutations, including EGFR and KRAS, is needed to guide effective treatment in patients with advanced NSCLC.

Clinical Outcomes in Hospitalized Patients with Cancer and New versus Preexistent Atrial Fibrillation.

There is limited information on the prognostic impact of new onset versus preexistent atrial fibrillation (AF) in hospitalized patients with cancer. We sought to determine the clinical impact of new onset AF (NOAF) compared with preexistent AF in hospitalized patients with cancer. All patients with cancer hospitalized over the course of one year with clinically manifest new or preexistent AF were enrolled in the Memorial Sloan Kettering Cancer Center (MSKCC) AF registry. The relationship of NOAF to the primary composite outcome of all cause death, cardiovascular (CV) rehospitalization or cerebrovascular event (CVE), as well as secondary CV endpoints, were analyzed using proportional hazards regression. Where applicable, the competing risk of death was accounted for using methodology described by Fine and Gray. Among 606 patients included in the analysis, 313 (51.7%) had NOAF and 293 (48.3%) had preexistent AF. Patients with NOAF were younger and had less frequent prior history of CV disease compared with patients with preexistent AF. At follow up, patients with NOAF had a higher adjusted hazard for the primary composite outcome versus patients with prior AF (HR 1.64, 95% CI 1.27, 2.13, p=0.002), as well as the secondary CV composite outcome of clinical AF recurrence, CV death, CV rehospitalization or CVE (HR 2.17, 95% CI 1.57, 2.99, P<0.0001). In hospitalized patients with cancer and electrocardiographically manifest new versus preexistent AF, NOAF was associated with a higher risk for the primary composite outcome of all-cause death, CV rehospitalization or CVE.

C-reactive protein levels, the prognostic nutritional index, and the lactate dehydrogenase-to-lymphocyte ratio are important prognostic factors in primary central nervous system lymphoma: a single-center study of 223 patients

Primary central nervous system lymphoma (PCNSL) is a rare and highly aggressive type of extranodal non-Hodgkin lymphoma (NHL), and the prognosis is poor. Currently, the most used prognostic models are the Memorial Sloan-Kettering Cancer Center (MSKCC) and International Extranodal Lymphoma Study Group (IELSG) scores; however, their predictive effects are changing with increasing incidence and changing treatment regimens. A growing body of evidence has demonstrated that inflammatory and nutritional markers are factors that can determine tumor prognosis. Therefore, the aim of this study was to identify and validate novel prognostic factors for PCNSL. Clinical information was collected from 223 patients with PCNSL. Patients younger than 18 years of age were excluded. Progression-free survival (PFS) and overall survival (OS) were used as endpoints, and receiver operating characteristic (ROC) curve analyses were conducted to determine the cutoff values for the inflammatory indicators. Correlations between variables and PFS or OS were assessed using univariate and multivariate analyses, and positive indicators were selected for survival analysis. A prognostic nutritional index (PNI) < 49.38 was associated with worse PFS (p = 0.003), and outcomes significantly differed between patients with a PNI ≥ 49.38 and < 49.38 (p < 0.001). Age < 60 years (p < 0.001) and C-reactive protein (CRP) levels < 3.14 (p = 0.001) were associated with better OS. In elderly patients (≥ 60 years), a lactate dehydrogenase-to-lymphocyte ratio (LLR) < 95.69 (p = 0.021) was associated with better OS, and the outcome significantly differed between patients with an LLR ≥ 95.69 and LLR < 95.69 (p = 0.015). The PNI and CRP levels are prognostic factors for PCNSL, and CRP was the first time shown to be a prognosis factor of PCNSL. In elderly patients with PCNSL, the LLR can predict prognosis.

Abstract IA050: Biomarker benchmarking across ancestral populations

Abstract Background The US FDA approved pembrolizumab for the treatment of patients (pts) with unresectable or metastatic tumors classified as TMB-high (⩾ 10 variants/Mb) and who have progressed on prior therapy. The degree to which TMB-high (TMB-H) generalizes as a biomarker in diverse ancestral populations remains unknown. Methods Using 3 clinical cohorts from the TCGA, Dana-Farber Cancer Institute (DFCI), and Memorial Sloan Kettering Cancer Center (MSKCC), genetic ancestry was determined for pts diagnosed with solid tumors. For pts in the TCGA cohort, TMB (mutations/Mb) was calculated using 2 methods: 1) Filtering out germline variants using paired normal tissue (gold standard) 2) Using publicly available reference panels to filter out germline variants, which in turn mirrors tumor-only TMB calling at DFCI. TCGA data was used to compute ancestry-specific calibration coefficients for panel tumor-only TMB with TMB paired (tumor/normal) as a benchmark. Ancestry-specific and cancer-specific TMB coefficients from TCGA were then projected onto the TMB estimates in the DFCI and MSKCC cohorts. Overall survival (OS) was estimated using the Kaplan-Meier method. Results In the TCGA cohort (N=3618 pts), germline contamination was associated with TMB (tumor-only) inflation across ancestral populations. TMB inflation was more pronounced in non-Europeans (non-EUR; 2.2 fold inflation) compared to EUR (1.5 fold inflation). We then computed calibration coefficients from the TCGA (methods) and projected them to the DFCI cohort (N=8,193) sequenced by a tumor-only panel. We computed raw TMB and calibrated TMB (TMB-c) for each of the pts at DFCI. Non-EUR with TMB-H tumors had significantly higher rates of false TMB-H (i.e., raw TMB-H corrected to TMB-c low) compared with EUR. Of 100 EUR, 21 would have false TMB-H versus 37 Asians and 44 Blacks. Among pts with non-small cell lung cancer treated with ICIs, those misclassified as TMB-high from tumor-only panels in DFCI and MSKCC cohorts did not associate with improved outcomes comapred to true TMB-high tumors. In the DFCI and MSKCC cohorts, higher TMB was associated with significantly longer OS in EUR but this was not the case in Blacks and Asians, although sample sizes were small. Conclusion: TMB estimates from tumor-only panels overclassified individuals into the TMB-high group due to germline contamination, and this bias was more pronounced in pts with Asian/African ancestry. Misclassifed TMB-H was associated with suboptimal survival outcomes compared to true TMB-H. Ancestry-aware tumor-only TMB calibration and ancestry-diverse biomarker studies are critical to ensure that existing disparities are not exacerbated in precision medicine. Citation Format: Amin Nassar, Elio Adib, Sarah Abou Alaiwi, Talal El Zarif, Stefan Groha, Elie W. Akl, Pier Vitale Nuzzo, Tarek H. Mouhieddine, Tomin Perea-Chamblee, Kodi Taraszka, Habib El-Khoury, Muhieddine Labban, Christopher Fong, Kanika S. Arora, Chris Labaki, Wenxin Xu, Guru P. Sonpavde, Robert I. Haddad, Kent W. Mouw, Marios Giannakis, Stephen Hodi, Noah Zaitlen, Adam Schoenfeld, Nikolaus Schultz, Michael F. Berger, Laura E. MacConaill, Guruprasad Ananda, David J. Kwiatkowski, Toni K. Choueiri, Deborah Schrag, Jian-Carrot Zhang, Alexander Gusev. Biomarker benchmarking across ancestral populations [abstract]. In: Proceedings of the 16th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2023 Sep 29-Oct 2;Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2023;32(12 Suppl):Abstract nr IA050.

Cutaneous T-Cell Lymphoma and Dupilumab Use: A Retrospective Matched Cohort Study of Clinical Characteristics and Treatment Outcomes

Introduction: Dupilumab is a human monoclonal antibody that inhibits signaling of IL-4 and IL-13 and is currently approved by the Food and Drug Administration for the treatment of atopic dermatitis (AD). While causality has not been established, sporadic reports have emerged associating dupilumab use with the onset, progression, or subsequent diagnosis of cutaneous T-cell lymphomas, in particular mycoses fungoides (MF). The natural history of MF diagnosed after dupilumab exposure remains unclear. We performed a retrospective matched cohort study to compare the clinical characteristics and treatment outcomes of patients with MF based on prior dupilumab exposure. Methods: Through the Memorial Sloan Kettering (MSK) Research &amp; Technology data delivery platform (DataLine), we conducted a search for any patient seen by a provider in the MSK Cutaneous Lymphoma Clinic whose electronic record contained the word ‘dupilumab’ or ‘Dupixent.‘ Once the initial query was performed, we individually reviewed each patient to confirm: (1) A histologically-confirmed diagnosis of MF by an MSK dermatopathologist, and (2) Use of dupilumab prior to the histological diagnosis of MF. After identifying a cohort of patients with confirmed MF after dupilumab use (cases), we identified a matched cohort of patients with confirmed MF and no prior dupilumab use (controls). Specifically, we matched patients for stage at diagnosis, date of birth and age at diagnosis (+/- 5 years), and date of presentation (+/- 5 years). Clinical data and treatment outcomes were collected by clinical review. Baseline characteristics were summarized using descriptive statistics and compared between the two groups using the Wilcoxon rank sum test, Pearson's Chi-squared test, and Fisher's exact test. Overall survival (OS), progression-free survival (PFS), and recurrence-free survival (RFS) were estimated as the elapsed time from initial diagnosis to death, disease recurrence, and progression, respectively. Kaplan-Meier estimates were stratified by exposure to dupilumab, and differences between the estimates were compared using the log-rank test. Results: We identified 25 patients who presented with a new, histologically-confirmed diagnosis of MF following dupilumab treatment for a diagnosis of AD. In 16 of these patients, AD was diagnosed based on a skin biopsy, whereas in 9 patients the diagnosis was based on clinical features only. Among 16 patients with known duration of dupilumab use, the median duration of use was 5.0 months (min/max: 1.5-24 months). In total, 14 patients were diagnosed with early-stage disease (IA n=2, IB n=12) and 11 were diagnosed with advanced stage disease (IIB n=4, IIIA n=2, IVA1 n=5). The estimated time from dupilumab treatment to lymphoma diagnosis was 10.2 months, and the median age at lymphoma diagnosis was 53 years (min/max: 28-84). Compared to the cohort of 25 matched patients with MF and no prior dupilumab use ( Table 1), no difference was observed in race, sex, initial therapy (skin-directed vs. systemic), or total number of therapies. The median follow-up time for patients exposed to dupilumab was 13.2 months (95% CI: 6.8-21.9), and 20.7 months (95% CI: 13.4-27.8) for those not exposed. No differences were observed in OS, PFS, or RFS between the exposed (dupilumab) and non-exposed (non-dupilumab) cohorts ( Figure 1). Two-year OS, PFS, and RFS in the non-dupilumab cohort was 85.0%, 47.3%, and 70.7%, compared to 100.0%, 41.5%, and 79.6% in the dupilumab cohort. One patient in the dupilumab-exposed cohort died of a non-lymphoma malignancy, whereas 4 patients in the non-exposed cohort died (3 due to lymphoma, 1 due to unknown causes). Conclusions: In summary, we identified a relatively large cohort of patients with MF diagnosed after dupilumab treatment. Patients presented with early and advanced-stage disease. Outcome were comparable to a matched cohort of non-exposed patients with MF. The causal relationship between dupilumab and MF remains unproven and deserves dedicated biological study.

Skeletal Muscle Index, As a Measure of Muscle Mass, Is a Significant Prognostic Factor for Newly Diagnosed Diffuse Large B-Cell Primary CNS Lymphoma Patients

Background: Primary cerebral lymphoma (PCNSL) is a rare subtype of diffuse large B-cell lymphoma (DLBCL) with a particularly poor outcome compared to systemic DLBCL. Several prognostic scores have been developed for PCNSL. These include the International Extranodal Lymphoma Study Group (IELSG) score with five prognostic factors (patient age, serum lactate dehydrogenase [LDH] level, Eastern Cooperative Oncology Group performance status [PS], deep brain involvement and cerebrospinal fluid [CSF] protein level), and the Memorial Sloan-Kettering Cancer Center (MSKCC) score, which is based on age and the Karnofsky Performance Score (KPS). However, the reproducibility of these two scoring systems is low, and the IESLG score requires CSF protein and LDH levels evaluation, which are not often available. None of these scores incorporate data on nutritional status, in particular albumin, which has been shown to be an important prognostic factor in systemic DLBCL. In addition, a recent study demonstrated that the muscle density and mass determined by imaging segmentation, were strong prognostic factors for newly diagnosed systemic DLBCL patients (Jullien et al. Cancers 2021). Our aim was to determine whether these imaging data were also predictive factors for outcome in newly diagnosed diffuse large B-cell PCNSL patients. Methods: For each patient included in the study, muscle mass and density were estimated using skeletal muscle index (SMI) and mean skeletal muscle radiation attenuation (SMRA), respectively. SMI and SMRA were computed at L3 level using the whole-body scan (WBS) performed at the diagnosis of PCNSL, by the mean of an automated segmenting algorithm. Muscle characteristics were then classified as normal or low using already validated cut-off according to patients' age and body mass index (Westenberg et al. Sci Rep 2023). Patients' outcomes were then compared based on their muscle quality, albumin level at diagnosis and other clinical variables commonly included in PCNSL prognostic scores. Results: Between 2015 and 2022, 98 consecutive diffuse large B-cell PCNSL patients were newly diagnosed at Hematology Department of Hospital Lyon Sud, France. Among them, 77 had WBS at diagnosis with available data for this study. The median age was 69 years (range 24-87) and 22 patients (29%) presented an altered ECOG PS (PS 3-4 score). All patients were treated with first-line high-dose methotrexate-based chemotherapy in combination with rituximab. SMI was normal in 64 patients (83%) and 17 patients (17%) had low SMI. Patients with low SMI had more frequently an altered PS (p=0.07) and had lower albumin level (mean 36.7 g/L vs 32.0 g/L, p=0.007). SMRA was normal in 30 patients (39%) and 47 patients (61%) had low SMRA. Patients with low SMRA were older (60 vs 70 years old, p=0.001). In univariate analysis, patients with low SMI had worse outcome with significantly shorter overall survival (p=0.002, median OS, 21.4 months vs not reached) and a trend towards shorter event-free survival (p=0.055, median EFS, 11.9 months vs 42.3 months, Figure 1). No significant prognostic difference was seen for SMRA for EFS or OS. The multivariate analysis showed that SMI (Hazard ratio [HR]=3.33, 95% confidence interval [95%CI]= 1.21-9.20, p=0.02) and age (HR =1.86, 95%CI =1.15-3.0, p=0.012) were strongly associated with OS but not tumor localization (HR =1.46, 95%CI =0.60-3.55, p=0.4), PS (HR = 1.54, 95%CI =0.65-3.65, p=0.32) and albumin level (HR =1.05, 95%CI =0.47-2.32, p=0.91). Age was significantly associated with EFS (HR=1.4, p = 0.04) with a more limited association of SMI with EFS (HR=2.03, p=0.095). In exploratory subgroup analyses, we showed that SMI still influenced OS in patients with PS 4 (10 months vs not reached, p= 0.002), suggesting that SMI may be a more appropriate prognostic factor than PS, especially for patients with motor deficit for whom PS is difficult to estimate. Conclusions: This retrospective analysis of newly diagnosed PCNSL patients showed a significant association between decreased muscle mass, measured by SMI and overall survival, after controlling for other prognostic variables. These findings warrant confirmative analyses to explore the relationship between muscle mass/density and diffuse large B-cell PCNSL patient's outcome and integrate SMI in PCNSL prognostic scores.

TP53 Y220C Mutations in Patients with Myeloid Malignancies

Background: TP53 gene mutations lead to inactivation of the protein p53, causing genome instability, oncogenesis, and treatment resistance conferring an unfavorable prognosis. TP53 Y220C missense mutations in exon 6 (p.659A&amp;gt;G) are described as a frequent ‘hot spot’ mutation in solid tumors. A novel Y220C-targeted small molecule (PC14586) has shown preliminary efficacy and safety in patients with solid tumors (Dumbrava, E., ASCO 2022). No data has been reported about TP53 Y220C in hematologic malignancies to date. We set out to evaluate the clinicopathologic characteristics of TP53 Y220C in hematologic malignancies and particularly, myeloid neoplasms. Methods: We identified all patients with hematologic neoplasms with TP53 Y220C mutations at University of Texas MD Anderson Cancer Center and Memorial Sloan-Kettering Cancer Center, detected by standard next generation sequencing assays, and focused on patients with myeloid malignancies; summarizing the patient characteristics, disease features and outcomes for patients with this specific TP53 mutation. Overall survival (OS) was assessed by the Kaplan-Meier method from date of diagnosis to date of death from any cause or date of last follow-up. Leukemia-free survival (LFS) was calculated from time to diagnosis of TP53 Y220C mutated myelodysplastic syndrome (MDS) to date of transformation to acute myeloid leukemia (AML) or death. Survival analysis was conducted only for newly diagnosed patients with available follow-up data. Results: We identified 143 patients (pts) with hematologic malignancies harboring a TP53 Y220C mutation. MDS and AML were the most frequent diagnoses (MDS, n=66; AML, n=48). Median age was 69 years (range, 32-92) and 56% of patients were male (Table 1). We then analyzed in more detail patients with MNs, including CHIP/CCUS, MDS, AML and MDS/MPN. Among those, in 76 (64%) pts the TP53 Y220C mutation was detected at initial diagnosis and in 43 (36%) pts, TP53 Y220C was identified in the relapsed/refractory setting. Therapy-related myeloid neoplasms (MNs) accounted for 51 of 119 pts (41%). Complex karyotype was frequent (n=88/119, 74%) in the setting of TP53 Y220C. Median TP53 Y220C variant allele frequency (VAF) was 32% (range, 1%-96%). Eighty-two patients (69%) had additional co-mutations: 24 (20%) pts had additional (non-Y220C) TP53 mutations only; 34 (29%) pts had only non- TP53 co-mutations (most frequently DNMT3A and TET2) and 24 (20%) pts had additional TP53 mutations in the setting of other co-mutations. With a median follow-up of 9 months, median OS for patients with newly diagnosed TP53 Y220C MDS (n=39) and AML (n=23) was 13.1 months (95% CI, 7.6 to 18.7) and 8 months (95% CI, 5.6 to 13.4), respectively ( p=0.06) (Figure 2). For the TP53 Y220C MDS cohort, AML transformation occurred in 13 (33%) patients, after a median of 14.3 months. Median LFS for these patients was 12.2 months (95% CI, 7.9 to 16.5). Modest and non-statistically significant differences in median LFS for MDS patients who did and did not receive stem cell transplant (SCT) were identified; MDS+SCT (n=12) 14.5 months (95% CI, 4 to 25) vs MDS+no-SCT (n=27) 9.3 months (95% CI, 4.6 to 14), p=0.18. Interestingly, four patients diagnosed with TP53 Y220C mutated CHIP/CCUS with a median VAF of 2% were identified, and with a median follow-up of 34.6 months, none have developed dysplastic changes or progression to MDS or AML at the time of cut off. Additional outcomes data will be forthcoming in the analysis of this cohort of patients with TP53 Y220C mutated myeloid neoplasms. Conclusions: TP53 Y220C mutations are present in malignant blood disorders and are particularly more common in myelodysplastic syndromes and acute myeloid leukemia, with associated poor outcomes. Novel targeted therapies for this TP53 variant (Y220C) would be of interest for this patient population.

The Complex Structural Variant Chromothripsis Can be Defined on Targeted Sequencing Panels, Allowing Direct Clinical Translation in Order to Improve Multiple Myeloma Prognostication

Introduction: the complex structural variant (SV) chromothripsis is an independent predictor of progression-free (PFS) and overall survival (OS) in newly diagnosed multiple myeloma (NDMM); predictive in multivariate analysis when considering ISS, IgH translocations and TP53 status (Rustad BCD 2020). Chromothripsis is not included in current prognostic scores, despite a prevalence of 24% in NDMM. While gold-standard definition requires whole genome sequencing (WGS), we've previously shown that copy-number (CN) signatures are also able to accurately define chromothripsis in whole exome sequencing (WES, Maclachlan Nat Com 2021). As neither WGS nor WES are currently used in clinical medicine, we applied CN signatures to targeted sequencing, aiming for direct clinical translation of this key biological feature into clinical practice. Memorial Sloan Kettering (MSK) now offers the MSK-Heme-IMPACT targeted sequencing panel to all MM patients having a bone marrow biopsy (BMBx), as well as offering retrospective analysis for current patients having leftover DNA from prior BMBx. Methods: We analyzed 420 samples having either myTYPE or MSK-Heme-IMPACT targeted sequencing at MSK prior to July 2023. We excluded those with whole genome duplication, applied CN signatures, then for samples from NDMM performed multivariate analysis for PFS prediction by backwards stepwise regression. In samples with CN signatures predictive for chromothripsis and with DNA available from the same BMBx, we performed parallel WGS in order to check for chromothripsis. WGS sequencing was at 60-80x for tumor, 30-40x for normal, with analysis including Battenberg for CN analysis, with SvABA, BRASS and GRIDSS for SV calling. Additional parallel WGS is in progress for 45 samples with NDMM prior to DVRd or DKRd, in order to examine chromothripsis prediction and PFS impact in the context of highly effective induction. Results: From 420 samples, 48 were collected at the precursor stage, 205 with NDMM, and 167 with relapsed disease. 39 patients had 2 assays performed, while the remainder were unique patient samples. Within NDMM, the range of induction regimens received included lenalidomide and dexamethasone with bortezomib (VRd), carfilzomib (KRd) or daratumumab (DRd), along with quadruplet therapy (DVRd or DKRd). Targeted sequencing data detected all 6 key CN features observed in WGS and WES analysis. Similar to WES, there was a lower contribution from very high breakpoint counts per 10mB and longer lengths of oscillating CN segments, due to the lower resolution of targeted sequencing data. De novo extraction of CN signatures defined 1 signature containing multiple features consistent with chromothripsis; high breakpoint count per 10mB, more jumps between adjacent CN segments, longer lengths of oscillating CN segments and a predominance of small CN segments. When examining samples from NDMM, contribution from this signature was predictive of PFS in multivariate analysis when considering age, ISS, t(4;14), TP53 status and gain1q21 (hazard ratio &amp;gt; 4, p=0.003). 33 samples had a predictive probability for chromothripsis &amp;gt; 0.6 (the cut-off in WGS / WES having the highest sensitivity and specificity for prediction), with also having further DNA available. 30/33 (91%) were confirmed to harbor chromothripsis on manual inspection of WGS data. 19/30 (63%) with chromothripsis were BMBx from NDMM, while 11/30 (37%) were from relapsed disease. Discussion: the complex SV chromothripsis is a key genomic risk factor predicting for poor PFS and OS in MM, which is currently not being captured in routine practice or in clinical trials. Here, we demonstrate that chromothripsis can be predicted from targeted sequencing data, confirmed by gold-standard WGS analysis. Ongoing studies are expanding our cohort of NDMM with concurrent targeting sequencing and WGS, to define the sensitivity and specificity of targeted-sequencing-based CN signature prediction of chromothripsis. The additional cohort largely comprises patients treated with daratumumab-quadruplet induction therapy, aiming to confirm that chromothripsis and CN signatures retain prognostic significance in the context of this therapy. With this data is in hand, we will develop a CN signature fitting tool, similar to mmsig for mutational signatures (Rustad Comm Biol 2021), to allow easily accessible incorporation of chromothripsis assessment into clinical practice.

Clinical Outcomes Associated with Del(17p) in Newly Diagnosed Multiple Myeloma Treated with Triplet and Daratumumab-Based Quadruplet Induction Regimens

Background: del(17p) is an independent prognostic marker associated with poor clinical outcome identified in a number of studies and has been incorporated into risk stratification models for multiple myeloma (Palumbo A et al J Clin Oncol 2015). Recent studies have reported that the percentage of myeloma cells with del(17p) by fluorescent in situ hybridization (FISH) can have a significant impact on outcomes (Corre J et al. Blood 2021; Thakurta A et al. Blood 2019; Thanendrarajan S et al. Haematologica 2017). Different collaborative groups have recommended different cut-points for the presence of del(17p) with clinical significance ranging from ≥20% or &amp;gt;55% of affected cells. Herein, we examined clinical outcomes of newly diagnosed multiple myeloma (NDMM) patients with del(17p) treated with novel combination triplet and quadruplet induction regimens, including bortezomib, lenalidomide, dexamethasone (VRd), carfilzomib, lenalidomide, dexamethasone (KRd), +/- daratumumab (DVRd and DKRd). Methods: We conducted a chart review study of 578 NDMM patients treated with VRd, KRd, DVRd and DKRd at Memorial Sloan Kettering (MSK) from 1/1/2015 to 5/1/23. Cutoff date for analysis was 7/19/23. Only patients with available cytogenetic results (FISH and/or SNP-array) were included. Patients who received ≤1 prior cycle of another regimen for MM were included. We analyzed the percentage of cells with del(17p) by inter-phase FISH analyses and/or SNP-array with CD138 enrichment when FISH was not available. A cut-point of 20% for del(17p) was used (del(17p) &amp;lt;20% vs del(17p) ≥20%) and compared with patients with standard-risk disease. Discrete patient characteristics were summarized by frequency (percentage) and continuous characteristics were summarized by median (Interquartile Range, IQR). PFS and OS were evaluated by Kaplan-Meier method. Results: There were 391 patients treated with triplet combination, of which 46 patients had del(17p) on baseline bone marrow biopsy. Among these patients, the presence of del(17p) in &amp;lt;20% of cells was detected in 13 patients, and the presence of del(17p) in ≥20% of affected cells was detected in 30 patients. Three patients did not have measurements. There were 187 patients treated with quadruplet combination, of which 20 had del(17p). Among these patients, 9 had del(17p) &amp;lt;20% of cells, 10 had del(17p) in ≥20% of cells, and one patient did not have measurements. There were 212 standard-risk NDMM patients treated with triplet induction and 99 patients treated with quadruplet regimens. The median age was 65 (IQR 54-73) for patients with del(17p) &amp;lt;20%, 61 (IQR 56-68) for del(17p) ≥20%, and 64 (IQR 56-69) for standard-risk patients (P=0.53). Other patient's baseline characteristics are summarized in Table 1. Best ORR among the three groups after triplet or quadruplet induction regimens were high: 95% for patients with del(17p) &amp;lt;20%, 93% for patients with del(17p) ≥20%, and 90% for standard-risk patients (P=0.53). After a median follow-up of 56.9 months, 2-year estimated PFS was 76% (95%CI, 59%-97%), 69% (95%CI, 55%-86%), and 85% (95%CI, 81%-90%) for del(17p) &amp;lt;20%, del(17p) ≥20%, and standard-risk groups, respectively. Median PFS was 48.9 months (95%CI, 26.9-not reached [NR]) for patients with del(17p) &amp;lt;20%, 34.3 months (29.8-66.7) for del(17p) ≥20%, and not reached (72.7-NR) for patients with standard-risk NDMM (P&amp;lt;0.0001). Median PFS was not significantly different between patients with del(17p) &amp;lt;20% and del(17p) ≥20% treated with triplet and quadruplet induction regimens (P=0.33). 2-year OS was 89% (95%CI 76%-100%), 85% (74%-98%), and 95% (93%-98%) for patients with del(17p) &amp;lt;20%, del(17p) ≥20%, and standard-risk (P=0.043). Conclusion: In the MSK experience, the presence of del(17p) was associated with poorer outcomes compared to standard-risk disease in patients with NDMM treated with both triplet and dara-based quadruplet induction regimens. We did not detect a significant difference in PFS and OS amongst del(17p) patients in our cohort using a cut-point of ≥20% or &amp;lt;20% of cells. These findings suggest that the presence of del(17p) alone portend poor prognosis, regardless of FISH cutoff threshold utilized. Data on TP53 mutation status and longer follow-up outcomes will be presented at the meeting.

Daratumumab, Bortezomib, Lenalidomide, and Dexamethasone (DVRd) and Daratumumab, Carfilzomib, Lenalidomide and Dexamethasone (DKRd) As Induction Therapy in Newly Diagnosed Multiple Myeloma

Background: Daratumumab, bortezomib, lenalidomide, dexamethasone (DVRd) and daratumumab, carfilzomib, lenalidomide, dexamethasone (DKRd) are promising induction regimens for newly diagnosed multiple myeloma (NDMM), resulting in deep responses and high minimal residual disease (MRD) negativity rates. Our prior work demonstrated progression-free survival (PFS) benefit in patients with high-risk (HR) NDMM receiving KRd versus VRd. Herein, we examined early outcomes associated with DVRd and DKRd induction in the management of patients with NDMM. Methods: We conducted a chart review study with NDMM patients treated with DVRd (N=107) and compared outcomes with patients treated with DKRd (N=82), of which 68 were on study (NCT03290950) at Memorial Sloan Kettering (MSK) from 10/5/2017 to 5/1/23. Cutoff date for analysis was 7/19/23. Patients who received ≤1 prior cycle of another regimen for MM were included (N=29). Patients who have not completed induction and received &amp;lt;4 cycles of a quadruplet regimen or who received the majority of their quadruplet induction outside of MSK were excluded. Bone marrow biopsies (BMBx) were typically performed for MRD evaluation prior to stem cell collection or at the end of cycle 8. MRD negativity was evaluated by flow cytometry at 10 -5 sensitivity threshold. Discrete patient characteristics were summarized by frequency (percentage) and continuous characteristics were summarized by median (Interquartile Range, IQR). PFS and overall survival (OS) were evaluated by Kaplan-Meier method. Logistic regression was used to estimate odds ratios (ORs) with 95% confidence intervals (CIs). Results: Median age was 66 (IQR, 60-70) for DVRd- and 59 (50-65) for DKRd-treated patients (P&amp;lt;0.001). In the DVRd group, 72% were White and 15% Black while the DKRd group had 79% were White and 9% Black. Most of the patients in both groups had RISS Stage 2 disease (DVRd RISS 1/2/3: 33%/61%/7%; DKRd: 35%/60%/5%). With HR cytogenetic abnormality (HRCA) defined as +1q, del(1p), t(4;14), t(14;16), t(14;20), and/or del(17p), 43/100 (43%) in DVRd group and 46/81 (57%) in the DKRd group had cytogenetic abnormalities that met HR criteria (P=0.065). Among the DVRd patients, 57, 28, and 15 had 0, 1, 2+ HRCA, respectively, while in the DKRd group, 35, 30, and 16 had 0, 1, 2+ HRCA, respectively (P=0.2). Median number of cycles was 6 (IQR, 4-6) for DVRd and 8 (IQR, 6-8) for DKRd (P&amp;lt;0.001). At data cutoff, 72 DVRd and 74 DKRd patients completed stem cell collection; among these patients, 41 (57%) DVRd- and 25 (34%) DKRd-treated patients received upfront ASCT. Best overall response rate (ORR) was 98% and 100% for DVRd and DKRd (P=0.5), respectively. Within 8 cycles of therapy, 21 (20%) and 28 (34%) patients achieved a ≥complete response (CR) in the DVRd and DKRd groups, respectively (P=0.024). Additionally, 83 (78%) patients were in ≥VGPR in the DVRd group and 74 (90%) in the DKRd group (P=0.021). There were 2 pts with best response as stable disease in the DVRd group. 86 DVRd pts and 73 DKRd patients had a BMBx within 8 cycles of induction therapy and were evaluable for MRD. Among these patients, 20 (23%) and 38 (52%) achieved MRD negativity for DVRd and DKRd, respectively. Multivariable analysis with clinical variables including type of quadruplet therapy, age, gender, race, R-ISS stage, cytogenetic risk, and number of cycles, demonstrated no significant association with achievement of CR (DKRd [ref] vs DVRd, OR 0.46, 95%CI 0.19-1.04; P=0.064). After a median follow-up of 13 (95%CI, 11-15) months for DVRd and 53 (48-55) months for DKRd, 1-year estimated PFS was 92% (95%CI, 86%-99%) and 94% (89%-99%) for DVRd and DKRd, respectively (P=0.5). Conclusion: In this single-center chart review study, DVRd was compared to DKRd as induction therapy for patients with NDMM. Importantly, DKRd patients were primarily comprised of patients from a clinical trial (NCT03290950), perhaps affecting demographics of the groups (DVRd group had older patients and more Black patients; DKRd group had more cycles of therapy due to trial design). DVRd and DKRd are both associated with high ORR. Although the ≥CR rate was greater with DKRd, on multivariable analysis there was no statistically significant difference between DKRd and DVRd after adjusting for various clinical variables. Data with follow-up outcomes will be presented at the meeting.

1429. Temperature and Humidity Variations in the Operating Room: A Risk Assessment

Abstract Background The American Society of Heating, Refrigeration, and Air-Conditioning Engineers (ASHRAE) recommends that hospital operating rooms (OR) are controlled for temperature and humidity with established ranges but does allow for exceedances for surgeons and surgical procedures. There is currently no clear consensus or established thresholds for out-of-range temperature and humidity ranges in relation to risk for surgical site infections (SSI). A study was performed at Memorial Sloan Kettering Cancer Center (MSKCC) to further examine if temperature and humidity deviations pose a patient safety risk. Methods The study included 5600 surgical procedures that were performed at MSKCC between September 2021 and January 2022 where temperature and humidity logs were available. Procedures were broken into groups based on in- (n=4419) and out-of-range (n=1187) temperatures and in- (n=5225) and out-of-range (n=381) humidity. SSIs following the procedure were identified using targeted positive microbiology cultures and within NHSN (National Healthcare Safety Network) surveillance periods. Chi-square and Fisher’s exact statistical tests were used to determine if there was a significant difference between groups where p &amp;lt; 0.05. Breakdown of eligible procedures and follow up Table represents the total number of eligible procedures with environmental parameters and SSI follow up period. Results There was no significant impact on SSI for out-of-range humidity (p=0.15). There was a significantly higher SSI rate observed for out-of-range temperatures (p=0.04) more specifically low out- of-range temperatures (p=0.02). A sub-analysis was performed for low temperatures based on procedures by duration. Significance of the finding was lost when surgery duration was 4 hours or greater. Conclusion If there was a true correlation between SSI and low ambient temperature, we would have expected to see a strengthening of SSI risk with longer procedures. This study suggests that temperature and humidity deviations out of the recommended guidelines in the operating room do not pose additional risks for surgical site infections and limited exceedances can be permitted without risking patient safety. Disclosures All Authors: No reported disclosures

2579. Pneumonia Severity Index and CURB-65 scoring for the evaluation of community associated pneumonia in patients with cancer at a major cancer center

Abstract Background Cancer patients with pneumonia (PNA) have 2 to 21 times higher mortality rate compared to the general population. The Pneumonia Severity Index (PSI) and CURB-65 scale have been used to predict mortality for patients with community acquired pneumonia (CAP). While these scales are well studied in the general population, there is a paucity of research on the utility of PSI and CURB-65 for predicting outcomes from CAP in patients with cancer. In this study, we assessed the relationship between PSI and CURB-65 score and 30-day mortality of CAP in cancer patients at Memorial Sloan Kettering Cancer Center (MSKCC). Methods A prospective study at MSKCC was conducted between 01/2023 and 06/2023 on patients who presented with clinical symptoms consistent with PNA and had a urine streptococcus antigen test sent for evaluation. PNA was defined as pulmonary infiltrates on chest X-ray or CT &amp;lt; 48 hours of index event and ≥ 1 clinical symptoms (fever, cough, dyspnea, increased sputum production, hypoxemia). Descriptive statistics were performed to report participant outcomes based on PSI and CURB-65 score and to identify the factors associated with CAP in this population. PSI and CURB-65 scores were derived as described in the literature, based on patient comorbidities (including neoplasia), clinical signs and laboratory values. Higher score is associated with worsening 30-day mortality. Results For 86 participants with who presented with symptoms of PNA and had a urine antigen test sent, 43 (50%) were diagnosed with CAP (Figure 1). The median age was 70 (range: 29-88), 21 (49%) were male, and 16 (37%) had hematologic malignancies (Table 1). Etiologies of CAP were identified in 8 patients (bacteria in 4 patients, virus in 3 patients, fungus in 1 patient). Of the 28 patients with 30-day outcome data available, 5 (18%) died within 30 days of CAP diagnosis. All 5 deceased participants were categorized as PSI Class IV or V (moderate to high risk); however, these participants only scored 0-2 (low to moderate risk) on the CURB-65 scale (Table 2). Conclusion We observed a 30-day overall mortality (n=28) of 18%. For cancer patients who present with CAP, PSI class provides a better means to assess mortality risk than CURB-65 score. Further study is needed to validate our findings. Disclosures Yeon Joo Lee, MD, MPH, AiCuris: institutional research support for clinical trials|Karius: institutional research support for clinical trials|Merck: Grant/Research Support|Scynexis: institutional research support for clinical trials Susan K. Seo, MD, Merck: Grant/Research Support

Multi-Institutional Development and Validation of a Radiomic Model to Predict Prostate Cancer Recurrence Following Radical Prostatectomy

The use of multiparametric magnetic resonance imaging (mpMRI)-derived radiomics has the potential to offer noninvasive, imaging-based biomarkers for the identification of subvisual characteristics indicative of a poor oncologic outcome. The present study, therefore, seeks to develop, validate, and assess the performance of an MRI-derived radiomic model for the prediction of prostate cancer (PC) recurrence following radical prostatectomy (RP) with curative intent. mpMRI imaging was obtained from 251 patients who had undergone an RP for the treatment of localized prostate cancer across two institutions and three surgeons. All patients had a minimum of 2 years follow-up via prostate-specific antigen serum testing. Each prostate mpMRI was individually reviewed, and the prostate was delineated as a single slice (ROI) on axial T2 high-resolution image sets. A total of 924 radiomic features were extracted and tested for stability via intraclass correlation coefficient (ICC) following image normalization via histogram matching. Fourteen important and nonredundant features were found to be predictors of PC recurrence at a mean ± SD of 3.2 ± 2.2 years post-RP. Five-fold, ten-run cross-validation of the model containing these fourteen features yielded an area under the curve (AUC) of 0.89 ± 0.04 in the training set (n = 225). In comparison, the University of California San Fransisco Cancer of the Prostate Risk Assessment score (UCSF-CAPRA) and Memorial Sloan Kettering Cancer Center (MSKCC) Pre-Radical prostatectomy nomograms yielded AUC of 0.66 ± 0.05 and 0.67 ± 0.05, respectively (p &lt; 0.01). When the radiomic model was applied to the test set (n = 26), AUC was 0.78; sensitivity, specificity, positive predictive value, and negative predictive value were 60%, 86%, 52%, and 89%, respectively. Accuracy in predicting PC recurrence was 81%.

Benefit and harm of lymphadenectomy in intermediate risk prostate cancer: comparison of five nomograms

BackgroundPelvic lymph node dissection (PLND) is recommended method for detecting prostate cancer (PCa) nodal metastases although associated with serious complications. In this study, we aimed to assess benefit/harm of routine PLND in intermediate risk PCa patients and to compare diagnostic yield of five different nomograms in predicting lymph node invasion (LNI).MethodsRetrospective analysis of consecutive PCa patients with intermediate risk of biochemical recurrence who underwent open radical prostatectomy (RP) with bilateral PLND between January 2017 and December 2019 at our institution. Partin, 2012-Briganti, 2018-Briganti, Cagiannos and Memorial Sloan Kettering Cancer Center (MSKCC) values were calculated. To compare accuracy, sensitivity, specificity, and area under receiver-operating curve (AUC) were calculated and then optimal cutoff values were estimated, analyses repeated and compared. To assess benefit and harm of PLND, relative risk (RR) and number need to treat (NNT) with LNI and complications set as outcome were calculated.ResultsTotal 309 subjects. Average age 62.2 years, average PSA 7.2 ng/mL; 18 (5.8%) had LNI; 88 (28.5%) suffered Clavien-Dindo grade 3–5 complication. AUC for predicting LNI: 0.729 for 2012-Briganti, 0.660 for MSKCC, 0.521 for 2018-Briganti, 0.486 for Cagiannos, and 0.424 for Partin. None of pairwise AUC comparisons based on default and newly established cutoff values were statistically significant. Lowest NNT was for Partin and Cagiannos with default cutoff (≥ 5%). Risks of serious complications between higher/lower than cutoff values were non-significant across nomograms.Conclusions2012-Briganti nomogram outperforms, although not significantly, MSKCC, 2018-Briganti, Cagiannos, and Partin nomograms in classifying LNI in intermediate risk PCa patients. Routine PLND in these patients should be avoided, due to high rate and severity of complications.

Genetic profile of Chinese patients with small bowel cancer categorized by anatomic location

BackgroundSmall bowel cancer (SBC) is a very rare solid malignancy. Consequently, compared with other malignant gastrointestinal tumors, our knowledge regarding SBC, specifically its molecular attributes, remains limited. Herein, we aim to provide an overview of the gene characteristics of Chinese patients with SBC, We particularly focus on elucidating the genetic intricacies that differentiate SBC patients whose primary tumors originate in distinct anatomical regions within the small bowel.MethodsDuring the period ranging from February 2018 to December 2022, a total of 298 tumor samples were consecutively collected from Chinese patients diagnosed with small bowel cancer.. Next-generation sequencing (NGS) was performed to detect gene mutation, assess microsatellite instability (MSI), and evaluate tumor mutational burden (TMB). Additionally,, IHC was used to analyze the level of PD-L1 expression within the samples.ResultsThe outcomes of the next-generation sequencing (NGS) unveiled the predominant gene mutations observed in Chinese patients with small bowel cancer (SBC). The top ten gene mutations identified were as follows: TP53 (53%), KRAS (51%), APC (31%), SMAD4 (19%), VEGFA (15%), CDKN2A (15%), RAC1 (15%), LRP1B (14%), MGMT (14%, CD74 (13%). Subsequent analysis revealed disparities in the gene landscape between the cohort in this study and that of the Memorial Sloan Kettering Cancer Center (MSKCC), Notably, distinguishable mutational frequencies were identified in several genes, including ERBB2, FBXW7, PIK3CA, etc. which exhibited contrasting presence in both this cohort and the MSKCC cohort.. Furthermore, we noticed variations in the frequency of gene mutations among SBC patients depending on the specific anatomical site where the tumors originated within the small bowel. In addition, the distribution of patients with high microsatellite instability (MSI-H) and tumor mutational burden (TMB) levels varied among SBC patients with tumors originating from the duodenum, jejunum, and ileum.ConclusionChinese patients with small bowel cancer exhibited a distinct genetic profile in comparison to other populations, highlighting a unique genetic landscape. Furthermore, noticeable disparities in the genetic landscape were observed between patients with cancer situated in the duodenum and those with cancer affecting other regions of the small bowel, this suggests that these patients should be treated differently.

EPCO-01. THE GENOMIC LANDSCAPE OF BRAIN METASTASIS

Abstract BACKGROUND The prognosis for patients with brain metastasis (BM) is devastating, while the underlying biology of BM development remains poorly understood. Identification of genomic biomarkers can offer promising opportunities for prediction of BM development and guidance in personalized treatment. In this study we evaluate the genomic alterations present in a large cohort of resected BM. METHODS Upon retrospective review, we identified 868 patients with diagnosis of solid primary cancers who had undergone a standard of care craniotomy at Memorial Sloan Kettering Cancer Center (MSKCC). BM samples were profiled by MSK-IMPACT, a next-generation sequencing (NGS) assay designed to detect a wide range of genetic alterations in 341-505 cancer genes. Genomic alterations were filtered for driver variants using OncoKB. Disease sites included cancers of head and neck, breast, lung, lower and upper gastrointestinal tract, kidney, ovary, uterus, prostate, skin, soft tissues, and others. RESULTS More than half (57%; 493/868) of patients were female and the median age was 63 (range 22-93). Foreseeably, the most common histology was non-small cell lung cancer (NSCLC; 38%), followed by invasive carcinoma of the breast and melanoma (16% and 13%, respectively). The most frequently encountered alterations were inTP53 (60%), CDKN2A (25%), TERT (23%), KRAS (18%), and PTEN (12%) across all BM samples. The median fraction of genome altered was 0.41 [range: 0-0.99] and the median tumor mutational burden was 6.6 muts/Mb [range: 0-395]. Additionally, TP53 was the most frequently altered gene after stratifying by primary histology in most cancer types, except for melanoma, renal, and thyroid, which were enriched for alterations in TERT, CDKN2A, and TERT, respectively. CONCLUSION The landscape of genomic alterations present in BM is distinct and varies by primary histologic diagnosis. Ongoing analyses of matched primary-BM pairs and clinicogenomic correlation will identify factors predisposing patients to BM development and progression.

PATH-37. PROGNOSTIC IMPACT OF TUMOR MUTATIONAL BURDEN AT BASELINE IN IDH-WILDTYPE GLIOBLASTOMA

Abstract Tumor mutational burden (TMB) has emerged as a potential biomarker for survival and response to immune checkpoint therapy. However, limited data are available regarding the prognostic impact of TMB at baseline in glioblastoma (GBM). This study aimed to evaluate the somatic mutation status and survival outcomes of isocitrate dehydrogenase (IDH) wildtype GBM using a dataset of GBM samples from cBioPortal and single tertiary hospital. A total of 840 samples were extracted from cBioPortal, including 410 samples from Memorial Sloan Kettering Cancer Center (MSKCC), 234 samples from The Cancer Genome Atlas (TCGA), 106 samples from Glioma Longitudinal Analysis Consortium (GLASS), and 90 samples from Clinical Proteomic Tumor Analysis Consortium (CPTAC). Additionally, we enrolled 341 GBM patients with targeted sequencing data (OncoPanel AMC) from Asan Medical Center. The TMBHigh group was defined as the top 10% value of TMB within each cohort. Among the total of 1,081 IDH-wt GBM samples, 104 samples were classified as TMBHigh. No significance difference in overall survival (OS) was observed between the TMBHigh group and the non-TMBHigh group across the 5 cohorts. The mutations of DNA mismatch repair genes (MLH1, MSH2, MSH6, PMS2, and POLE) was significantly higher in the TMBHigh group compared with the non-TMBHigh group in 3 of 5 cohorts (pASAN=1.02x10-4, pMSKCC=1.5x10-2, and pCPTAC=1.26x10-2). However, when we calculated the indel to total mutation ratio (I-index) to dedect microsatellite Instability, only the MSKCC cohort showed a higher I-index in the TMBHigh group compared to the non-TMBHigh group (16.7% vs 13.7%, p=2.67x10-3). In summary, there is no significant association between TMB at baseline and poor survival outcomes in IDH-wildtype GBM. Further research is needed to explore the potential implications of TMB as a prognostic marker and its relevance for personalized treatment strategies in GBM patients.