Search

Tumor genotyping is not routinely performed in localized non–small cell lung cancer (NSCLC) due to lack of associations of mutations with outcome. Here, we analyze 232 consecutive patients with localized NSCLC and demonstrate that KEAP1 and NFE2L2 mutations are predictive of high rates of local recurrence (LR) after radiotherapy but not surgery. Half of LRs occurred in tumors with KEAP1/NFE2L2 mutations, indicating that they are major molecular drivers of clinical radioresistance. Next, we functionally evaluate KEAP1/NFE2L2 mutations in our radiotherapy cohort and demonstrate that only pathogenic mutations are associated with radioresistance. Furthermore, expression of NFE2L2 target genes does not predict LR, underscoring the utility of tumor genotyping. Finally, we show that glutaminase inhibition preferentially radiosensitizes KEAP1-mutant cells via depletion of glutathione and increased radiation-induced DNA damage. Our findings suggest that genotyping for KEAP1/NFE2L2 mutations could facilitate treatment personalization and provide a potential strategy for overcoming radioresistance conferred by these mutations.Significance:This study shows that mutations in KEAP1 and NFE2L2 predict for LR after radiotherapy but not surgery in patients with NSCLC. Approximately half of all LRs are associated with these mutations and glutaminase inhibition may allow personalized radiosensitization of KEAP1/NFE2L2-mutant tumors.This article is highlighted in the In This Issue feature, p. 1775

Table S1 - S7

Supplementary Figures 1-9

Supplementary Figures related to Figures 1, 2, 3, and 4.

Table S2: Surveillance plasma sample ctDNA metrics. Table S3: Baseline plasma sample ctDNA metrics. Table S4: Surveillance plasma sample variants in ctDNA detected samples. Table S5: Baseline plasma variants in available samples. Table S6: Tumor variants detected by whole exome and targeted sequencing. Table S7: Canonical genes implicated in clonal hematopoiesis.

Table S1 shows summary patient demographics, pathologic features, and treatment details

Purpose:Treatment with PD-(L)1 blockade can produce remarkably durable responses in patients with non–small cell lung cancer (NSCLC). However, a significant fraction of long-term responders ultimately progress and predictors of late progression are unknown. We hypothesized that circulating tumor DNA (ctDNA) analysis of long-term responders to PD-(L)1 blockade may differentiate those who will achieve ongoing benefit from those at risk of eventual progression.Experimental Design:In patients with advanced NSCLC achieving long-term benefit from PD-(L)1 blockade (progression-free survival ≥ 12 months), plasma was collected at a surveillance timepoint late during/after treatment to interrogate ctDNA by Cancer Personalized Profiling by Deep Sequencing. Tumor tissue was available for 24 patients and was profiled by whole-exome sequencing (n = 18) or by targeted sequencing (n = 6).Results:Thirty-one patients with NSCLC with long-term benefit to PD-(L)1 blockade were identified, and ctDNA was analyzed in surveillance blood samples collected at a median of 26.7 months after initiation of therapy. Nine patients also had baseline plasma samples available, and all had detectable ctDNA prior to therapy initiation. At the surveillance timepoint, 27 patients had undetectable ctDNA and 25 (93%) have remained progression-free; in contrast, all 4 patients with detectable ctDNA eventually progressed [Fisher P < 0.0001; positive predictive value = 1, 95% confidence interval (CI), 0.51–1; negative predictive value = 0.93 (95% CI, 0.80–0.99)].Conclusions:ctDNA analysis can noninvasively identify minimal residual disease in patients with long-term responses to PD-(L)1 blockade and predict the risk of eventual progression. If validated, ctDNA surveillance may facilitate personalization of the duration of immune checkpoint blockade and enable early intervention in patients at high risk for progression.

Circulating tumor DNA (ctDNA) molecular residual disease (MRD) following curative-intent treatment strongly predicts recurrence in multiple tumor types, but whether further treatment can improve outcomes in patients with MRD remains unclear. We applied CAPP-Seq ctDNA analysis to 218 samples from 65 patients receiving chemoradiation therapy (CRT) for locally advanced NSCLC, including 28 patients receiving consolidation immune checkpoint inhibition (CICI). Patients with undetectable ctDNA after CRT had excellent outcomes whether or not they received CICI. Among such patients, one died from CICI-related pneumonitis, highlighting the potential utility of only treating patients with MRD. In contrast, patients with MRD after CRT who received CICI had significantly better outcomes than patients who did not receive CICI. Furthermore, the ctDNA response pattern early during CICI identified patients responding to consolidation therapy. Our results suggest that CICI improves outcomes for NSCLC patients with MRD and that ctDNA analysis may facilitate personalization of consolidation therapy.

Chromosomal rearrangements involving the mixed lineage leukemia (MLL) gene, also known as KMT2A, are often observed in human leukemias and are generally associated with a poor prognosis. To model these leukemias, we applied clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 gene editing to induce MLL chromosomal rearrangements in human hematopoietic stem and progenitor cells purified from umbilical cord blood. Electroporation of ribonucleoprotein complexes containing chemically modified synthetic single guide RNAs and purified Cas9 protein induced translocations between chromosomes 9 and 11 [t(9;11)] at an efficiency >1%. Transplantation of gene-edited cells into immune-compromised mice rapidly induced acute leukemias of different lineages and often with multiclonal origins dictated by the duration of in vitro culture prior to transplantation. Breakpoint junction sequences served as biomarkers to monitor clonal selection and progression in culture and in vivo. High-dimensional cell surface and intracellular protein analysis by mass cytometry (CyTOF) revealed that gene-edited leukemias recapitulated disease-specific protein expression observed in human patients and showed that MLL-rearranged (MLLr) mixed phenotype acute leukemias (MPALs) were more similar to acute myeloid leukemias (AMLs) than to acute lymphoblastic leukemias (ALLs). Therefore, highly efficient generation of MLL chromosomal translocations in primary human blood stem cells using CRISPR/Cas9 reliably models human acute MLLr leukemia and provides an experimental platform for basic and translational studies of leukemia biology and therapeutics.

Tumor genotyping is not routinely performed in localized non–small cell lung cancer (NSCLC) due to lack of associations of mutations with outcome. Here, we analyze 232 consecutive patients with localized NSCLC and demonstrate that KEAP1 and NFE2L2 mutations are predictive of high rates of local recurrence (LR) after radiotherapy but not surgery. Half of LRs occurred in tumors with KEAP1/NFE2L2 mutations, indicating that they are major molecular drivers of clinical radioresistance. Next, we functionally evaluate KEAP1/NFE2L2 mutations in our radiotherapy cohort and demonstrate that only pathogenic mutations are associated with radioresistance. Furthermore, expression of NFE2L2 target genes does not predict LR, underscoring the utility of tumor genotyping. Finally, we show that glutaminase inhibition preferentially radiosensitizes KEAP1-mutant cells via depletion of glutathione and increased radiation-induced DNA damage. Our findings suggest that genotyping for KEAP1/NFE2L2 mutations could facilitate treatment personalization and provide a potential strategy for overcoming radioresistance conferred by these mutations. Significance: This study shows that mutations in KEAP1 and NFE2L2 predict for LR after radiotherapy but not surgery in patients with NSCLC. Approximately half of all LRs are associated with these mutations and glutaminase inhibition may allow personalized radiosensitization of KEAP1/NFE2L2-mutant tumors. This article is highlighted in the In This Issue feature, p. 1775

Purpose: Treatment with PD-(L)1 blockade can produce remarkably durable responses in patients with non–small cell lung cancer (NSCLC). However, a significant fraction of long-term responders ultimately progress and predictors of late progression are unknown. We hypothesized that circulating tumor DNA (ctDNA) analysis of long-term responders to PD-(L)1 blockade may differentiate those who will achieve ongoing benefit from those at risk of eventual progression. Experimental Design: In patients with advanced NSCLC achieving long-term benefit from PD-(L)1 blockade (progression-free survival ≥ 12 months), plasma was collected at a surveillance timepoint late during/after treatment to interrogate ctDNA by Cancer Personalized Profiling by Deep Sequencing. Tumor tissue was available for 24 patients and was profiled by whole-exome sequencing (n = 18) or by targeted sequencing (n = 6). Results: Thirty-one patients with NSCLC with long-term benefit to PD-(L)1 blockade were identified, and ctDNA was analyzed in surveillance blood samples collected at a median of 26.7 months after initiation of therapy. Nine patients also had baseline plasma samples available, and all had detectable ctDNA prior to therapy initiation. At the surveillance timepoint, 27 patients had undetectable ctDNA and 25 (93%) have remained progression-free; in contrast, all 4 patients with detectable ctDNA eventually progressed [Fisher P < 0.0001; positive predictive value = 1, 95% confidence interval (CI), 0.51–1; negative predictive value = 0.93 (95% CI, 0.80–0.99)]. Conclusions: ctDNA analysis can noninvasively identify minimal residual disease in patients with long-term responses to PD-(L)1 blockade and predict the risk of eventual progression. If validated, ctDNA surveillance may facilitate personalization of the duration of immune checkpoint blockade and enable early intervention in patients at high risk for progression.

RNA sequencing is emerging as a powerful technique to detect a diverse array of fusions in human neoplasia, but few clinically validated assays have been described to date. We designed and validated a hybrid-capture RNAseq assay for FFPE tissue (Fusion-STAMP). It fully targets the transcript isoforms of 43 genes selected for their known impact as actionable targets of existing and emerging anti-cancer therapies (especially in lung adenocarcinomas), prognostic features, and/or utility as diagnostic cancer biomarkers (especially in sarcomas). 57 fusion results across 34 samples were evaluated. Fusion-STAMP demonstrated high overall accuracy with 98% sensitivity and 94% specificity for fusion detection. There was high intra- and inter-run reproducibility. Detection was sensitive to approximately 10% tumor, though this is expected to be impacted by fusion transcript expression levels, hybrid capture efficiency, and RNA quality. Challenges of clinically validating RNA sequencing for fusion detection include a low average RNA quality in FFPE specimens, and variable RNA total content and expression profile per cell. These challenges contribute to highly variable on-target rates, total read pairs, and total mapped read pairs. False positive results may be caused by intergenic splicing, barcode hopping / index hopping, or misalignment. Despite this, Fusion-STAMP demonstrates high overall performance metrics for qualitative fusion detection and is expected to provide clinical utility in identifying actionable fusions.

Summary Although treatment of non-small cell lung cancer (NSCLC) with immune checkpoint inhibitors (ICIs) can produce remarkably durable responses, most patients develop early disease progression. Furthermore, initial response assessment by conventional imaging is often unable to identify which patients will achieve durable clinical benefit (DCB). Here, we demonstrate that pre-treatment circulating tumor DNA (ctDNA) and peripheral CD8 T cell levels are independently associated with DCB. We further show that ctDNA dynamics after a single infusion can aid in identification of patients who will achieve DCB. Integrating these determinants, we developed and validated an entirely noninvasive multiparameter assay (DIREct-On, Durable Immunotherapy Response Estimation by immune profiling and ctDNA-On-treatment) that robustly predicts which patients will achieve DCB with higher accuracy than any individual feature. Taken together, these results demonstrate that integrated ctDNA and circulating immune cell profiling can provide accurate, noninvasive, and early forecasting of ultimate outcomes for NSCLC patients receiving ICIs.

There are many potential sources of error in clinical practice. An astute clinician must not only work hard to prevent errors but also minimize harmful sequelae that could arise from errors that do occur. A rare but real source of error is cross-contamination of pathology specimens. Such contaminants are colloquially referred to as floaters. If not recognized expediently, floaters can lead to misdiagnoses that may prompt unnecessary and inappropriate treatment. We report the case of a patient with a benign adnexal neoplasm on the face that, due to cross-contamination of pathology specimens, was initially diagnosed as an aggressive invasive melanoma that would have warranted wide local excision and sentinel lymph node biopsy.

FLT3 fusions are associated with myeloid and lymphoid neoplasms with eosinophilia. We describe a patient presenting with clinicopathologic features of both chronic eosinophilic leukemia, not otherwise specified (CEL, NOS) and systemic mastocytosis (SM). The bone marrow demonstrated a myeloproliferative neoplasm with eosinophilia and aggregates of atypical mast cells. Cytogenetic analysis revealed a t(13;14)(q12;q32), which was subsequently molecularly characterized as a novel TRIP11-FLT3 rearrangement. A KIT D816V mutation was also identified. The patient rapidly transformed to T-lymphoblastic leukemia/lymphoma and expired shortly after diagnosis. This is the fifth FLT3 fusion gene described in the literature; the presence of both myeloid and lymphoid neoplasms implicates involvement of an early hematopoietic progenitor by rearranged FLT3. We suggest that leukemias and lymphomas with FLT3 fusion genes exhibit similar clinicopathologic features to, and should be included in, the WHO category of "Myeloid and lymphoid neoplasms with eosinophilia and abnormalities of PDGFRA, PDGFRB, or FGFR1, or with PCM1-JAK2."

Background Next-generation sequencing (NGS) panels have created an unprecedented opportunity to interrogate a broad array of variants in a multiplex fashion. Few data have demonstrated how NGS testing impacts diagnosis and treatment decisions. In this retrospective analysis, we evaluated the clinical application of NGS myeloid panels (MP) in patients (pts) evaluated by physicians in the Stanford Division of Hematology. Methods The study was approved by the Stanford University IRB. The cohort consisted of 1,015 pts ≥18 years of age (median 66; range 18-96 years; 51% female). A total of 1,213 MPs obtained from peripheral blood (n=568) or bone marrow (n=645) from March 2017 to June 2018 were analyzed at Stanford (n=761) using the 54-gene TruSight® Myeloid Sequencing Panel (Illumina, San Diego, CA) or Oregon Health & Science University (n=452) using the GeneTrails® Hematologic Malignancies 76-Gene Panel. Electronic medical records were reviewed from t-3 to t+9 months from when MPs were obtained to assess physician reasoning for MP acquisition, documentation of results, and how results were clinically applied. We defined three categories of MP acquisition: 1) diagnostic clarification, 2) prognostication and/or management guidance, or 3) minimal residual disease (MRD) monitoring. We analyzed changes in clinical management, including addition of targeted or non-targeted therapeutics, clinical trial eligibility, or other practice recommendations. Results Of the 1,213 MPs, 882 (73%) demonstrated at least one pathogenic/ likely pathogenic variant (median 2; range 1-8). Median turn-around-time was 18 days (range 7-31) and average cost was $2,600. Of all MPs, 462 (38%) were obtained for diagnostic clarification, 732 (60%) for prognostication / management guidance in pts with known myeloid (n=701) or lymphoid (n=31) neoplasms, and 19 (2%) for MRD monitoring, although the assay was not designed for this indication. MPs were ordered to clarify a diagnosis for the following reasons: unexplained cytopenia(s) (n=199), molecular profiling for a suspected hematolymphoid neoplasm (n=156), unexplained -cytosis (n=86), and testing for other lab abnormalities (e.g. elevated serum tryptase, paraproteinemia) (n=18), or signs (e.g. splenomegaly, splenic vein thrombosis)(n=3). A pathogenic/likely pathogenic variant was found in 294 (64%) pts, confirming or establishing the presence of a myeloid (n=266) or lymphoid neoplasm (n=7) or resulting in a diagnosis of CHIP (n=9) or CCUS (n=12) in pts who did not meet 2016 World Health Organization diagnostic criteria for a hematolymphoid neoplasm (Fig. 1). Of the 732 MPs ordered for prognostication/ management guidance, 272 MPs (37%) were obtained in the initial workup of non-APL AML pts. The frequency of favorable (21%), intermediate (55%), and adverse risk (24%) genetics according to the European LeukemiaNet stratification in non-APL AML is shown in Figure 2a, which also denotes the frequency of favorable (9%) or adverse risk (70%) variants in MDS, and adverse risk variants in MF (52%), MDS/MPN (68%), and advanced systemic mastocytosis (44%). Among MPs obtained for prognostication/ management guidance, 163 (22%) led to a modification in clinical practice, divided between 132 (18%) which led the physician to change therapy (e.g. FDA-approved targeted therapy, clinical trial, or FDA-approved therapy, such as hypomethylating agents in MDS) and 31 (4%) resulted in a non-therapeutic change (e.g. expedited HSCT referral or more frequent follow-up)(Fig. 2b). 87 pts with a myeloid neoplasm had 184 repeat MPs for relapsed/refractory disease (n=52), transformation to higher-risk MDS or AML (n=45) or progressive cytopenias (n=87). Among these MPs, 38 (21%) identified a new pathogenic/likely pathogenic variant of which 29% (n=11) led to either a) initiation of targeted therapy with enasidenib in relapsed IDH2+ AML (n=3) or midostaurin for secondary FLT3+ AML from MDS (n=3), or b) consideration for a clinical trial with a splicing modulator for MDS characterized by a splicing variant (n=5). Conclusion In our academic hematology practice, two-thirds of MPs ordered for diagnostic clarification identified a pathogenic/likely pathogenetic variant that helped to confirm or establish a new diagnosis of a hematolymphoid neoplasm, CHIP, or CCUS. In addition, approximately 20% of MPs ordered for prognostication/ management guidance led to a change in clinical practice. Disclosures Gotlib: Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Blueprint Medicines: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Promedior: Research Funding; Pharmacyclics: Research Funding; Gilead: Membership on an entity's Board of Directors or advisory committees, Research Funding; Allakos: Honoraria, Membership on an entity's Board of Directors or advisory committees; Deceiphera: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Seattle Genetics: Research Funding; Incyte: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.

In order to identify novel somatic mutations associated with classic BCR/ABL1-negative myeloproliferative neoplasms, we performed high-coverage genome sequencing of DNA from peripheral blood granulocytes and cultured skin fibroblasts from a patient with MPL W515K-positive primary myelofibrosis. The primary myelofibrosis genome had a low somatic mutation rate, consistent with that observed in similar hematopoietic tumor genomes. Interfacing of whole-genome DNA sequence data with RNA expression data identified three somatic mutations of potential functional significance: i) a nonsense mutation in CARD6, implicated in modulation of NF-kappaB activation; ii) a 19-base pair deletion involving a potential regulatory region in the 5'-untranslated region of BRD2, implicated in transcriptional regulation and cell cycle control; and iii) a non-synonymous point mutation in KIAA0355, an uncharacterized protein. Additional mutations in three genes (CAP2, SOX30, and MFRP) were also evident, albeit with no support for expression at the RNA level. Re-sequencing of these six genes in 178 patients with polycythemia vera, essential thrombocythemia, and myelofibrosis did not identify recurrent somatic mutations in these genes. Finally, we describe methods for reducing false-positive variant calls in the analysis of hematologic malignancies with a low somatic mutation rate. This trial is registered with ClinicalTrials.gov (NCT01108159).

We demonstrate the feasibility of using glass microbeads with a holographic barcode identifier to track DNA specimens in the molecular pathology laboratory. These beads can be added to peripheral blood specimens and are carried through automated DNA extraction protocols that use magnetic glass particles. We found that an adequate number of microbeads are consistently carried over during genomic DNA extraction to allow specimen identification, that the beads do not interfere with the performance of several different molecular assays, and that the beads and genomic DNA remain stable when stored together under regular storage conditions in the molecular pathology laboratory. The beads function as an internal, easily readable specimen barcode. This approach may be useful for identifying DNA specimens and reducing errors associated with molecular laboratory testing.

Abstract 2863 Background: JAK2 V617F is a recurrent, activating mutation in patients (pts) with BCR-ABL1-negative MPNs. Mutations in codon 515 of MPL occur in 1–5% and 5–10% of ET and PMF pts, respectively, and similar to JAK2 V617F, lead to constitutive JAK-STAT signaling. The acquisition of multiple mutational events affecting the JAK-STAT axis or components of the epigenetic machinery is common in MPNs and likely contributes to phenotypic diversity, including progression to acute myeloid leukemia. Mutated genes thus far implicated in MPN initiation and/or progression include TET2, CBL, SH2B3, ASXL1, DNMT3A, IDH1/2, IKZF1, EZH2, SRSF2, and TP53. In order to identify novel somatic mutations associated with classic BCR-ABL1-negative MPNs, we performed whole genome sequencing of DNA extracted from peripheral blood granulocytes and cultured skin fibroblasts from a patient with PMF and a known MPL W515K mutation. Methods: Whole genome sequencing (WGS) was undertaken in a 55 year-old man with untreated PMF four years after initial diagnosis. His DIPSS Plus risk group was low (score 0). His karyotype was normal, and molecular testing revealed wild-type JAK2 in addition to the MPL W515K mutation. WGS of purified granulocytes and paired cultured skin fibroblasts was performed using both Illumina HiSeq and Complete Genomics (CGI) platforms. The resulting data were analyzed using multiple independent aligners and variant callers. Amplicon-based targeted resequencing with the Illumina MiSeq platform was used to evaluate additional patient samples for recurrent mutations. Stanford institutional review board approval and informed pt consent was obtained for these analyses. Results: The PMF genome was sequenced to 88X (Illumina) and 128X (CGI) average fold coverage, and the cultured skin fibroblast genome was sequenced to 47X (Illumina) and 126X (CGI). The PMF genome had a low somatic mutation rate, consistent with that observed for other sequenced hematopoietic tumor genomes, with a low number of true somatic mutation calls. To definitively identify true mutations among various sequencing artefacts and germline variants, we use cultured skin fibroblasts which can be prepared with no contamination by neoplastic cells. In addition to re-identification of the MPL W515K mutation, this approach identified six additional somatic mutations that alter gene coding regions, splice sites, or known regulatory regions: a nonsense mutation in CARD6, implicated in modulation of NF-kappaB activation; a splice-site mutation in CAP2; three nonsynonymous point mutations in KIAA0355, SOX30, MFRP; and a 19-base pair (bp) deletion involving a regulatory region in the 5'-untranslated region (5'-UTR) of BRD2, a bromodomain-containing protein implicated in transcriptional regulation (Table). CARD6, BRD2, and KIAA0355, an uncharacterized protein, are expressed by the granulocytes derived from this patient, supporting a potential role in the development of PMF in this pt. Using massively parallel sequencing, we are currently examining the transcribed region of BRD2 and the coding region of the other five genes in additional pt samples. Analysis of the first 87 samples (MF=47, PV=20, ET=20) out of a cohort of 180 MPN pts has thus far not identified recurrent somatic mutations in these six genes. Conclusion: High-coverage genome sequencing of neoplastic and germline cells from a patient with MPL-mutated PMF identified six additional somatic mutations of potential functional significance. Work is ongoing to determine if somatic mutations in these genes are found in other patients with BCR-ABL1-negative MPNs, and their pathogenetic relevance to MPN biology. Disclosures: Snyder: Illumina: Consultancy; GenapSys: Membership on an entity's Board of Directors or advisory committees; Personalis: Consultancy, Founder Other.