Maximilian Diehn, Ph.D. - US grants

DESCRIPTION (provided by applicant): Early detection of cancer via screening has been shown to lead to improved survival for several common malignancies. However, screening comes with significant risks and expenses. One important issue faced by all screening tests is detecting as many cancers as possible while minimizing identification of false positives. False positive screening results induce anxiety in patients and their families, require additional expensive tests, and may result in harm if a follow-up study leads to a complication. We propose to develop a novel, genomic approach for cancer screening that leverages insights gained from high throughput re-sequencing of cancer genomes. We will develop this method in the context of lung cancer, since it is the number one cause of cancer deaths and since low dose computed tomography (CT) screening has recently been shown to produce significant survival benefits in high-risk patients. However, ~95% of positive screening results from low dose CT lung cancer screenings are false positives and so improvements are clearly needed. This proposal describes our plan to develop, optimize, and test our method. We will perform both pre-clinical and clinical evaluations and will test our approach in multiple settings, includig as a secondary screening procedure for differentiating between true positive and false positive screening results and as a primary screening modality. Importantly, the method is readily extendable to any cancer for which high throughout sequencing data are available and we envision ultimately being able to screen for most common cancers using a single assay.

? DESCRIPTION (provided by applicant): Radiotherapy plays a critical role in the definitive management of early stage and locally advanced non- small cell lung cancer (NSCLC), the number one cause of cancer deaths in the United States. Two important challenges in the management of NSCLC patients treated with radiotherapy are the inability to detect microscopic residual disease and the difficulty of distinguishing between recurrent malignancy and post-radiation normal tissue changes on follow-up radiographic studies. These shortcomings prevent design and testing of individualized treatment strategies that could potentially increase cure rates and suggest that improved methods for following patients after treatment are needed. Our goal is to test a novel approach for detecting the presence of residual or recurrent disease in NSCLC patients treated with radiotherapy. Somatic alterations in the genomes of cancer cells represent ideal biomarkers, since they can define patient- and tumor-specific signatures that distinguish cancer cells from non-malignant counterparts. Cancers continually release genomic DNA into the circulation, potentially allowing noninvasive access to cancer genomes and quantitation of disease burden. We have designed and implemented a novel, high throughput sequencing-based technique for measurement of circulating tumor-derived DNA termed CAPP- Seq (CAncer Personalized Profiling in by deep Sequencing). We propose to evaluate the utility of ctDNA analysis via CAPP-Seq in NSCLC patients treated with radiotherapy and to explore mechanisms of ctDNA release and clearance in human patients. Taken together, these studies will lay the groundwork for prospective clinical trials that will test personalization of radiotherpy-based therapeutic strategies using ctDNA analysis.

SUMMARY Our proposed research is aimed at developing catalytic molecules and methods that greatly improve the recovery of biomolecular information from formalin-fixed tissue specimens. Formalin (formaldehyde) treatment of tissue is universally used in preparation of hundreds of millions of biopsy and surgery specimens worldwide. Unfortunately, the formaldehyde forms many adducts and crosslinks with the biomolecules in the specimens, strongly inhibiting the ability to obtain sequences and quantification of RNAs and DNAs from such specimens, information that is increasingly essential to diagnosis and treatment of cancer. Standard heating-based methods of RNA/DNA extraction remove only a fraction of adducts, and they are harsh, damaging the nucleic acids in the process. In our phase I work we demonstrated that the use of catalytic methods could greatly enhance the removal of formaldehyde adducts from DNA and RNA bases. Importantly, we were able to develop catalytic protocols that function successfully with actual formalin-fixed specimens. Amplifiable RNA was recovered in amounts as much as 25-fold greater than a world-leading commercial protocol, an unprecedented magnitude of improvement. The goals of this collaborative phase II research are to move to practical development of this technology for application in clinically important applications with multiple tissue types. We will further optimize our catalytic protocols for DNA recovery and minimization of damage, for application in next- generation sequencing (NGS). We will test whether new catalyst structures can be yet more effective than early examples, and we will examine whether optimal RNA recovery results in improved representation of gene expression and microRNA expression by NGS and microarray analysis. Finally, we will test application in lung tumor needle biopsies, which can be difficult to analyze by current methods because of the small amount of material present. This research program is innovative because it represents the first concept of using catalysts to reverse formaldehyde adducts of biomolecules, and because new, 2nd-generation catalysts and protocols are planned. The work we propose is significant because FFPE methodology is the universal worldwide standard for storing tissue specimens during cancer diagnosis. Approximately 400 million existing specimens could benefit from success of the work, and millions of new patients could be positively affected annually. We expect that our catalytic methods will become part of the new standard of practice for molecular diagnosis and therapy.

PROJECT SUMMARY/ABSTRACT PIs: Ash Alizadeh, M.D./Ph.D. & Maximilian Diehn, M.D./Ph.D. For patients with Diffuse large B-cell lymphoma (DLBCL), the most common lymphoma subtype, curative outcomes are common. Unfortunately, despite many large clinical trials, survival has not significantly improved over the last 15 years and nearly a third of patients continue to succumb to this disease. For these patients, effective strategies to predict early treatment failures have been elusive. Our long-term goal is to study the ability of baseline and dynamic risk factors, including genetic mutations and circulating tumor DNA (ctDNA), to accurately predict treatment outcomes in DLBCL patients. Our central hypothesis is that novel biomarkers of cancer risk, such as detection of ctDNA and detailed genetic profiling, can be used for early detection of residual disease, to identify dynamic changes that anticipate treatment failure, and to provide early surrogate endpoints for future clinical trials. We will test our hypothesis via three specific aims: (1) To build an accurate and dynamic predictor of survival for patients newly diagnosed with DLBCL, (2) To test the validity and utility of this predictor in a large multi-institutional cohort of patients from around the globe, and (3) To assess the ability of this dynamic risk assessment tool to serve as an early surrogate endpoint in prospective clinical trials. We will apply our novel approach in both the frontline and relapse/refractory setting and to a variety of treatment types including immunochemotherapy, an antibody-drug conjugate and Chimeric Antigen Receptor (CAR) T cells. If successful, our project will lead to novel ways to select better therapies for patients at highest risk of failure. Our innovative approach, in which we will employ novel, blood-based methods for tumor genotyping and disease monitoring that were developed by our group, will lay the foundation for studies aimed at reducing risk of treatment failure in DLBCL patients. Demonstrating that this approach can serve as a robust, early surrogate endpoint for patients with aggressive lymphomas would be transformative for future trial design and for rapid evaluation of novel, personalized treatment approaches in patients at highest risk for recurrence. Our work will serve as proof-of-principle for an approach that could also be applied to other cancer types.

ABSTRACT Non-small cell lung cancer (NSCLC) is a major disease burden in the United States and worldwide. Most patients are diagnosed at an advanced stage. For unresectable locally advanced NSCLC, the standard of care is definitive concurrent chemoradiotherapy. Unfortunately, the majority of patients will develop local-regional or distant failure with standard treatment. High-dose radiotherapy or consolidation chemotherapy may reduce local or distant recurrence, but are also associated with significant toxicity leading to morbidity and even mortality. Several randomized phase III trials failed to show a survival benefit with intensified treatment given to unselected, locally advanced NSCLC populations, highlighting the limitations of current `one-size-fits-all' treatment. A biomarker-driven approach would allow rational treatment selection based on individualized assessment of risks of local-regional versus distant failure. However, current imaging and genomic markers lack sufficient accuracy in predicting relevant outcomes. The goal of this project is to develop and validate quantitative imaging biomarkers to evaluate early response and integrate with circulating tumor DNA analysis to predict patterns of treatment failure in locally advanced NSCLC. Previously, we developed a novel tumor partitioning method based on FDG-PET and CT images, which revealed spatially distinct tumor subregions with predictive significance in NSCLC. In this project, we will further improve our tumor partitioning method to identify robust subregions, and propose novel image features to characterize intratumoral spatial heterogeneity via spatially explicit analysis. A rigorous qualification procedure will be employed to identify repeatable and reproducible image features for biomarker discovery. We will develop a predictive imaging biomarker by incorporating pre and mid-treatment scans in a retrospective patient cohort, and independently test it in two prospectively collected cohorts including a national randomized phase II trial. Finally, we will combine imaging with circulating tumor DNA analysis in a unifying model to further improve predictive accuracy. We anticipate that the integrated biomarker will allow reliable, early prediction of local-regional vs distant failure, which has important implications for deciding treatment between high-dose RT vs intensive systemic therapy. If successful, the proposed biomarkers will afford a rational approach to individualized therapy and ultimately improve outcomes in locally advanced NSCLC.

PROJECT SUMMARY Bladder cancer (BC) is the sixth most common cancer in the U.S., has one of the highest recurrence rates of all solid cancers, and is the most expensive cancer to treat from diagnosis to death. There are significant unmet needs for biomarkers and molecular diagnostic tools to better inform decision making across all stages of BC. This includes prediction of treatment response in patients with non-muscle invasive (NMIBC) and muscle invasive bladder cancer (MIBC), as well as improving diagnostic yield in patients undergoing screening cystoscopy for hematuria. Our long-term goal is to improve outcomes for BC patients through the development and application of molecular biomarkers that facilitate personalized approaches to detection and treatment. Urine is an attractive source for development of BC diagnostics and we recently developed a novel strategy for detecting urine tumor DNA called urine tumor DNA Cancer Personalized Profiling by Deep Sequencing (uCAPP-Seq). Our preliminary data indicate that uCAPP-Seq has outstanding sensitivity and specificity for detection and surveillance of BC. In this project we will prospectively collect urine and other biospecimens from patients with or at risk for BC and will test the potential clinical utility of uCAPP-Seq in different clinical scenarios. We will also test if augmenting uCAPP-Seq with analysis of urinary RNA or DNA methylation further augments performance. Our central hypothesis is that uCAPP-Seq will enable monitoring of BC responses during and after treatment for NMIBC and MIBC. Furthermore, we hypothesize that combining analysis of urine DNA mutations, DNA methylation, and urine RNA will allow ultrasensitive and specific early detection of BC. We propose three specific aims: 1) To assess the value of urine tumor DNA for non-invasive response assessment and monitoring in patients with high risk NMIBC treated with bacillus Calmette-Guérin (BCG) immunotherapy; 2) To determine if urine tumor DNA analysis can predict pathologic complete responses to neoadjuvant chemotherapy in patients with MIBC; and 3) To develop a Bladder Cancer Interception Assay (BCIA) which integrates utDNA mutations and methylation with urinary RNA for ultra-sensitive detection of bladder cancer. Successful completion of the studies proposed here will serve as a foundation for incorporating our novel urine-based biomarkers into prospective clinical trials. We foresee that our approach will allow personalization of treatment strategies to improve outcomes for BC patients. Importnatly, our work will serve as proof-of- principle for an approach that could also be applied to other genitourinary cancer types.

PROJECT SUMMARY/ABSTRACT PIs: Maximilian Diehn, M.D./Ph.D. & Ash Alizadeh, M.D./Ph.D. Non-small cell lung cancer (NSCLC) is the most common cancer in the U.S. and the number one cause of cancer-related deaths. Radiation therapy (RT) plays a critical role in the treatment of NSCLC, both in the curative and palliative settings. While advances in tumor imaging and radiation delivery techniques over the past several decades have significantly improved RT, advances in genomic and molecular understanding of tumors have largely failed to impact management of patients treated with RT. Therefore, development of ?precision radiation oncology? approaches, defined as the use of molecular biomarkers to personalize RT, remains a major unmet need. Additionally, predicting which patients will develop RT-induced toxicity remains a challenge and prevents early intervention prior to onset of symptoms. Our long-term goal is to develop novel, molecularly-based precision radiation oncology approaches for NSCLC patients treated with RT. Our central hypothesis is that novel biomarkers of recurrence risk, such as analysis of ctDNA and genetic profiling, can be used for early prediction of treatment outcomes while a patient is still on therapy. We will test our hypothesis via three specific aims: (1) To establish the ability of mid-treatment ctDNA changes to predict ultimate outcomes in locally advanced NSCLC patients treated with RT, (2) To develop novel, personalized risk models that integrate molecular and clinical factors and can accurately predict the risk of recurrence, and (3) To test the hypothesis that a novel liquid biopsy approach we have recently developed can predict which patients will develop symptomatic radiation pneumonitis. If successful, our project will lead to novel ways to personalize therapy for locally advanced NSCLC patients treated with RT. Our innovative approach, in which we will employ blood-based methods for tumor genotyping, disease monitoring, and toxicity prediction that were developed by our group, will lay the foundation for studies aimed at reducing risk of treatment failure and toxicity in NSCLC patients treated with RT. We envision that our approach will enable future trial designs that implement molecularly-driven precision radiation oncology and will facilitate treatment escalation for patients at highest risk of recurrence and de-escalation for those at lowest risk. Additionally, our work will serve as proof-of-principle for an approach that could also be applied to other areas of radiation oncology.

PROJECT SUMMARY/ABSTRACT PIs: Ash Alizadeh, M.D./Ph.D. & Maximilian Diehn, M.D./Ph.D. Classical Hodgkin lymphoma (HL) is among the most curable human malignancies. However, strategies to personalize HL therapies and to minimize long-term attendant toxicities of chemotherapy are currently limited to baseline risk factors and imaging. This is due to our incomplete understanding of targetable pathways and lack of good biomarkers. Because of the low fraction of malignant cells in tumor tissue and consecutive technical challenges, the landscape of HL is not well-defined. Our long-term goal is to study the ability of baseline and dynamic risk factors, including genetic mutations, circulating tumor DNA (ctDNA) and imaging studies (PET), to accurately predict treatment outcomes in HL patients, and to provide a basis for individualized precision medicine. Our central hypothesis is that clinical and biological heterogeneity in HL reflects distinct genomic features that are noninvasively measurable using ultrasensitive ctDNA techniques, and that refining early response assessment integrating interim PET and blood based methods improves prognostication. We will test our hypotheses via three specific aims: (1) To noninvasively define the genomic landscape of somatic variations in HL, and to determine the relationship of genomic variants with biological heterogeneity at initial disease presentation, (2) To associate molecular features at baseline and molecular response with ultimate therapeutic outcome, and to integrate clinical and molecular biomarkers in a personalized dynamic risk model for predicting HL outcomes, and (3) To functionally characterize novel mutations in Interleukin-4 receptor (IL4R) resulting in gain-of-function IL4/STAT6 signaling, and to test the utility of precision therapeutic targeting of these mutations. If successful, our project will lead to novel ways to select better therapies for patients at highest risk of failure, and to minimize toxicity for the majority of patients responding well to standard therapy. Our innovative approach, in which we will combine blood-based methods for genotyping and disease monitoring with imaging studies, will provide the basis for a personalized treatment approach in HL.