Nadauld: Have a pers. cancer medicine clinic; testing; molecular tumor board; drug procurement. Debate on in-house or outsource #AMP2015

8:19am November 6th 2015 via Hootsuite

Nadauld: About 5M participants, including WY, ID, TX; an 'open system' - facilities open to both MDs and patients. #AMP2015

8:18am November 6th 2015 via Hootsuite

Nadauld: We're now entering application of genomics on a larger scale, and measuring outcomes. Intermountain: payor/provider system #AMP2015

8:17am November 6th 2015 via Hootsuite

Nadauld: Can still apply genomic science, use molecular profiling with genes en-masse, a segmentation on a molecular basis for Rx #AMP2015

8:16am November 6th 2015 via Hootsuite

Nadauld: Actionable mutations per tumor - often only 2-4; most of mutations are passengers, essentially. Where the tumor board acts #AMP2015

8:15am November 6th 2015 via Hootsuite

Nadauld: Number of mutations by tumor type - 135 for melanoma, 147 for NSCLC; but # medications limited https://t.co/wtSF6cPh7P #AMP2015

8:14am November 6th 2015 via Hootsuite

Nadauld: The flip side: 'anticipation-based chemotherapy', assessing new alterations that evolve, and adjust Rx #AMP2015

8:13am November 6th 2015 via Hootsuite

Nadauld: Shows tumor evolution, from one set of mutations to another '13 Puente ref https://t.co/OidHnS16J1 #AMP2015

8:12am November 6th 2015 via Hootsuite

Nadauld: Ref Gerlinger '12 NEJM https://t.co/wt6mAkcrCm for heterogeneity, also figure from '13 Nature https://t.co/eJ3Ek762oC #AMP2015

8:11am November 6th 2015 via Hootsuite

Nadauld: 22 hospitals, 180 clinics: starts with a simple view (cell -> analysis -> list of var's -> line up Rx) but heterogeneity..

8:09am November 6th 2015 via Hootsuite

Lincoln Nadauld (Intermountain) "Role of a molecular tumor board in clinical cancer genomics" #AMP2015

8:05am November 6th 2015 via Hootsuite

ICYMI: QIAGEN GeneReader at #AMP2015 | Next Generation Technologist https://t.co/1PsDM29UP3

7:31am November 6th 2015 via Hootsuite

.@madduri Alas at the beginning of #AMP2015 the announcement was made not to take photos of slides, so I respect that request...

6:55am November 6th 2015 via Hootsuite in reply to madduri

RT @splon: Gordon Mills - personalized medicine is more stratified medicine. 8 subclasses of BRCA makes clinical trials difficult #targets15

8:55pm November 5th 2015 via Hootsuite

RT @kennamshaw: Gordon Mills, my boss, up next on Delivering the promise of molecular medicine #Targets15; long list of disclosures ha!

7:55pm November 5th 2015 via Hootsuite

RT @HarvardBiz: How Making Time for Books Made Me Feel Less Busy https://t.co/vhaJNpYPO0 https://t.co/4DxovIN2Wk

7:50pm November 5th 2015 via Hootsuite

Williams:A2: Correlative studies are needed, accepting proposals for future studies, will use it in the future #AMP2015

6:40pm November 5th 2015 via Hootsuite

Williams:Q:What about liquid biopsies? A:Very interested, we are collecting blood and banked. 27 companies - but standards needed #AMP2015

6:39pm November 5th 2015 via Hootsuite

Williams:A2:NIST NA12878 same pipeline will have 1% mismatch (!) be careful; easy to sequence / actionable #AMP2015

6:38pm November 5th 2015 via Hootsuite

Williams: Q:Different var callers? A:Diff SW, diff params will change results. Thus: they locked down Ion Torrent Suite 4.4 #AMP2015

6:38pm November 5th 2015 via Hootsuite

Williams: ;Acknowledges Jason Lih, David Sims from NCI. MATCH is 15 working groups, >100 involved. MDACC, Raja Luthra; MGH Iafrede #AMP20

6:35pm November 5th 2015 via Hootsuite

Williams: Concordance well-above 95%; LOD was 5% for SNVs, could get 'a little below that' but kept at 5%. CNV, feel it can improve #AMP2015

6:34pm November 5th 2015 via Hootsuite

Williams: Wide tissue distribution pie chart shown; specificity across all var types very high (100%); reprod across 16 smpls #AMP2015

6:33pm November 5th 2015 via Hootsuite

Williams: Reviews validation plan - 25 SNVs, 10 indels, 10 large indels, 10 CNV, 10 fusions. #AMP2015

6:29pm November 5th 2015 via Hootsuite

Williams: Showed very busy pie chart with 'hit rates' for distribution of mut/trtmnt. 'Nowhere near where we thought it would be' #AMP2015

6:28pm November 5th 2015 via Hootsuite

Williams: Filed intended use, validation plan, pre-submission to FDA, had to go to NY State, locked protocol #AMP2015

6:25pm November 5th 2015 via Hootsuite

Williams: Selected OncoMine Cancer Panel; 143 genes, 4066 var's. SNV, indel, CNV, trargeted translocation. MDA, MassGen, Yale, NCI #AMP2015

6:24pm November 5th 2015 via Hootsuite

Williams: MATCH has a goal of 25% of enrollment to be 'rare' tumors. Common is NSCLC, breast, colon, prostate #AMP2015

6:23pm November 5th 2015 via Hootsuite

Williams: Open to all solid tumors, lymphoma. Open to progression after at least 1 line of std therapy #AMP2015

6:23pm November 5th 2015 via Hootsuite

Williams: Onto MATCH, four seq labs. If a match, onto a treatment arm (no randomization). If pt progresses, w/in 6 mo onto another #AMP2015

6:22pm November 5th 2015 via Hootsuite

Williams: Each treatment arm is indep; showed map of the US and the cooperative group nationwide. 120 P3 and 215 early phase trials #AMP2015

6:21pm November 5th 2015 via Hootsuite

Williams:Open at up to 2,400 clinical sites (!), incl. community hospitals. 1 Master IND, central IRB, single clin database #AMP2015

6:19pm November 5th 2015 via Hootsuite

Williams: FDA greenlighted; 67% 'hit rate' to put pts on protocol, majority were P53 mutations. Joint NCI/NNCN trial #AMP2015

6:18pm November 5th 2015 via Hootsuite

Williams: Made 62 gene custom @iontorrent AmpliSeq panel; first 50pts reflexed to Sanger. After validation, went back to FDA #AMP2015

6:17pm November 5th 2015 via Hootsuite

Williams: Rules for actionable mutation - GoF (7 genes) LoF (13 genes); found 391 aMOI's (actionable mutation of interest) #AMP2015

6:16pm November 5th 2015 via Hootsuite

Williams: For mutations detected, two random arms, assigned to target mut or not. MPACT has 20 genes, 3 major pathways #AMP2015

6:13pm November 5th 2015 via Hootsuite

Williams: FDA rec'd: Difficult to sequence: sens/spec for homopolymer subst. and homopolymers. #AMP2015

6:12pm November 5th 2015 via Hootsuite

Williams: For MPACT: a 10-day window for assay results; >50% tumor nuclei; define sens/spec by variant type. Large indels >=3bp #AMP20

6:11pm November 5th 2015 via Hootsuite

Williams: Wrote up validation plan document, pre-submission discussion with FDA, assay intended use, risk. Informal FDA discussion AMP2015

6:09pm November 5th 2015 via Hootsuite

Williams: The FDA is very interested in LDTs, provided draft guidance; and asking for 'design control' and 'quality systems' #AMP2015

6:07pm November 5th 2015 via Hootsuite

Williams: Clinical Utility: does it provide useful clinical info? Does it improve clinical outcome of a patient? #AMP2015

6:06pm November 5th 2015 via Hootsuite

Williams: Two trials: MPACT and MATCH. Assay success: both analytical and clinical validity (pos and neg PPV) Then: clin utility #AMP2015

6:06pm November 5th 2015 via Hootsuite

Williams: Public data sharing of clinical outcome - even though pt confidentiality is a concern #AMP2015

6:05pm November 5th 2015 via Hootsuite

Williams: VUS as a community, we need to work together and share what they do, what protein does, how pts respond #AMP2015

6:04pm November 5th 2015 via Hootsuite

Williams: Data reporting stds such that others can understand assay bias, and repeat data. Public data sharing w/assay details #AMP2015

6:03pm November 5th 2015 via Hootsuite

Williams: NIST started the #GIAB (Genome in a Bottle): some regions everybody can get, other regions that no-one can sequence #AMP2015

6:02pm November 5th 2015 via Hootsuite

Williams: 'There is a need for assay standards'; guidance on what specific gene var's are actionable, ass'd levels of evidence #AMP2015

6:02pm November 5th 2015 via Hootsuite

Williams: Many different variable parameters that add bias; different cut points. Impt to remember. "Everybody is doing it" #AMP2015

6:01pm November 5th 2015 via Hootsuite

Williams: 'A tale of 2 studies': "I'll start with some of my biases" - powerful technology, NGS assays are complex #AMP2015

6:00pm November 5th 2015 via Hootsuite

P. Mickey Williams (NCI Frederick, MD) "Application of Targeted Next Generation Sequencing (NGS) in Clinical Oncology Studies" #AMP2015

5:59pm November 5th 2015 via Hootsuite