Search

Tumor Mutational Burden (TMB) is a promising biomarker for prediction of response to immune checkpoint blockade (ICB). It is uncertain whether ICB has prognostic value outside of ICB therapy. The CMS400 next generation sequencing panel (NGS) is a 409 gene, 15,992 amplicon, and 1.745 Mb panel instituted during 2014-2015 and run for 556 cancer patients who had been consented for participation within a prospective molecular pathology biomarker trial (PA14-0099). All patients had advanced or recurrent solid tumor malignancies that were refractory to at least one line of systemic therapy prior to enrollment. Survival time was calculated from time of NGS-tested tissue collection. TMB was calculated by dividing reported mutations (RM) by 1.745Mb, the genetic footprint of the NGS panel. Subtraction of germline single nucleotide polymorphisms was performed for each patient. GraphPad Prism 7.03 software was used to calculate p values and to plot Kaplan-Meier survival curves. One hundred seven patients (19.2%) received ICB. When stratified by reported mutations (RM: 0, 1, 2, 3, 4-5, 6-7, 8-9, 10-18, and >19), a statistically significant decrement of overall survival was seen with increasing TMB in patients not treated with ICB (Table 1, p Median Survival Stratified by Tumor Mutational Burden and ICB Treatment StatusAll PtsAll PtsICB TreatedICB TreatedNo ICBNo ICBReported Mutations# of PtsMedian Survival (Months)# of PtsMedian Survival (Months)# of PtsMedian Survival (Months)p-value (Log-rank)Hazard Ratio of ICB Therapy95% CI of HR of ICB Therapy0 RM7150.71958.85248.50.3190.7050.368 - 1.341 RM8550.91262.07347.40.3390.6840.345 - 1.362 RM9433.91928.07533.90.8650.9510.535 - 1.693 RM8830.51951.26928.90.1890.7040.428 - 1.164-5 RM9330.81141.08230.40.5390.8150.442 - 1.506-7 RM4928.41127.33828.550.6230.8270.401 - 1.718-18 RM5623.95828.84823.10.5491.2520.553 - 2.84>19 RM2041.38102.31224.20.00230.1800.0599 - 0.543Total55635.810747.944933.40.00490.7070.567 - 0.882p-value (Log-rank) Citation Format: Richard K. Yang, Peng Wang, Fatima Z. Jelloul, Mark J. Routbort, Scott Kopetz, Kenna R. Shaw, Jack J. Lee, Jiexin Zhang, Hui Chen, Keyur P. Patel, Raja Luthra, Russell R. Broaddus. Prognostic value of tumor mutational burden using a 409 gene NGS panel in cancer patients with advanced stage recurrent or treatment refractory disease [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3162.

Background: We developed and established a high depth targeted sequencing platform of 202 genes (all exons) to identify actionable DNA alterations in tumor samples. We aimed a minimum of 500x depth and defined actionable if the aberration had a direct link to an available clinical treatment or clinical trial. Our platform was optimized for FFPE specimens and low input DNA while the data analysis pipeline was designed to detect low frequency mutations and copy number alterations. Over 500 tumor samples and matched normal were analyzed using this platform. Methods: Libraries were made from a minimum of 170 ng of FFPE DNA (tumors) or blood DNA (matched normal). Capture of 202 genes (over 5,000 exons) was performed using Nimblegen probes to a minimum of 50x fold enrichment. Captured libraries were sequenced using Illumina HiSeq2000. Duplicate reads were removed from the raw data and the reads were mapped hg19 reference genome. We used VarScan2 for calling somatic, germline and loss of heterozygosity SNVs and short indels. Copy number variation was also reported when significant gain/loss on an exon was detected. Additionally, we used other tolls to annotate the functional consequence of the SNVs, such as CanDrA to predict if an SNV is driver or passenger event. Results: We sequenced over 500 tumor samples and their normal match (blood). The mean coverage for most of samples was over 1500x, which allowed us to detect low frequency mutations (5% or higher) with accuracy. The most common disease sites analyzed where we found genetic alterations were: breast, colon, brain, ovary, endometrial, skin, lung, prostate, head and neck, sarcoma, kidney and stomach. About 98% of all tumors analyzed had at least one somatic alteration (SNVs or copy number) and although the vast majority of the samples analyzed were derived from FFPE blocks, the number of samples that failed was less than 5%. Among the most common genes mutated are several “actionable genes” such as BRAF, EGFR and KRAS. The samples tested in this platform were also tested on ion torrent Ampliseq 46 gene panel (46 overlapping genes) in a CLIA certified laboratory at MD Anderson Cancer Center with a very high level of concordance. Conclusion: With a very low level of failure and high depth provided, our results indicate that this platform can be used for FFPE tumor samples with high level of accuracy and sensitivity, showing that this platform is reliable to be implemented in clinical settings. Citation Format: Karina Eterovic, Ken Chen, Hao Zhao, Funda Meric-Bernstam, Raja Luthra, Aldape Kenneth, Mark Routbort, Scott Kopetz, Michael Davies, John de Groot, Stacy Moulder, Yong Mao, Chacha Horombe, Lin-ya Tang, Song Ping, Zhang Qingxiu, Ezzeddine Nader, Lan Zhang, Kenna M. Shaw, John Mendelsohn, Gordon B. Mills. T200: a high depth targeted exome sequencing platform to identify actionable alterations in FFPE solid tumor samples. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4276. doi:10.1158/1538-7445.AM2014-4276

Objective: Clinically based risk assessment tools, targeting those with young age of cancer onset and family history of specific cancers, have been used to identify individuals with Lynch Syndrome (LS). Women with LS are equally likely to develop endometrial carcinoma (EC) as they are colorectal carcinoma (CRC). To identify women presenting with EC at risk for LS, the Society of Gynecologic Oncology (SGO) published recommendations in 2007 regarding which patients would benefit from further genetic evaluation for LS. These criteria are comparable to the Revised Bethesda Guidelines. The primary objective of this study was to evaluate SGO criteria's ability to predict women at risk for LS in the EC population and to ascertain if alternative criteria exist that can better identify high-risk patients. Methods: 408 sequential EC cases were evaluated for immunohistochemical expression of four DNA mismatch repair (MMR) proteins. Tumors with loss of MSH2, MSH6 or PMS2 were designated as probable Lynch Syndrome (PLS). Tumors with loss of MLH1 and absence of MLH1 promoter methylation were also designated PLS. Clinical and pathologic data were collected from the electronic medical record. Results: Of the 408 EC cases, 43 (10.5%) of the patients were defined as probable Lynch Syndrome (PLS). 97/408 (23.7%) of EC cases met SGO criteria, but only 14 of these 97 cases (14.4%) were PLS. Of the 43 PLS cases, 29/43 (67.4%) did not meet SGO criteria. Comparison of clinical and pathologic characteristics, including age of cancer diagnosis and family history of EC and/or CRC, revealed no statistically significant differences between sporadic and PLS tumors with the exception of tumors arising from the lower uterine segment. Tumors with this sight of origin are more likely to be associated with PLS. The sensitivity and specificity of SGO criteria was 32.6% and 77.2%, respectively. Conclusions: Existing clinical guidelines for detecting endometrial carcinoma patients at elevated risk for having Lynch Syndrome perform poorly in an unselected patient population. With the exception of tumors arising from the lower uterine segment, there are no statistically significant clinical or pathological differences between sporadic tumors and PLS tumors. Lower uterine segment EC occurs in only 3% of all EC cases and is not a useful screening tool. SGO clinical criteria correctly identified 32.6% of women with PLS, which results in missed CRC screening opportunities for 67% of women with increased risk profiles. These results suggest that the SGO guidelines are inadequate for identifying LS. Given the lack of effective clinical or pathological screening tools, MMR IHC and MLH1 methylation testing of all endometrial cancer patients may be the best way to identify women at risk for having Lynch Syndrome. Citation Format: Amanda S. Bruegl, Bojana Djordjevic, Bryan M. Fellman, Diana L. Urbauer, Raja Luthra, Karen H. Lu. Poor performance of published clinical screening criteria for the population-based identification of endometrial cancer patients with Lynch Syndrome. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4855. doi:10.1158/1538-7445.AM2013-4855

Purpose: C-MET is over-expressed in metastatic melanoma and have copy number gains at chromosome 7 in late stages of melanoma progression. However, the presence of C-MET mutation is not well known in melanoma. We analyzed tumor samples of patients with malignant melanoma to identify the frequency and the clinicopathology of C-MET mutations. Methods: We identified 103 patients with metastatic mutation who underwent testing for c-MET mutation and reviewed the clinical data of these patients. The sequencing analyses were performed by Sequenom Assay, and the mutations were confirmed with Sanger Sequencing analysis. Clinical characteristics were correlated with C-MET mutation status, and a survival analysis was performed to identify significant associations. Results Among the 103 patients, eleven (11%) patients had melanoma harboring a C-MET mutation. Five (4.5%) patients had N375S mutation; two (1.8%) patients had R988C mutation; two (1.8%) had T1010I; one (1%) patient had H1112R and another patient had N375/T1010 mutation. A median age at diagnosis, sex, race and the status of ulceration at the primary site and the stages at diagnosis were similar between patients with a C-MET mutation and those with wild type. There were no significant differences in the overall survival from the time of stage IV diagnosis between the two groups, but there was a trend of a shorter median duration from the time of the initial diagnosis to distant metastases among patients with a C-MET mutation. (23.6 months vs. 15.9 months, p=0.09). Conclusions We found C-MET mutations in 11% of patients with malignant melanoma. There was a tendency to have distant metastasis at an earlier time in patients with a C-MET mutation. This is the first report describing the frequency and clinical characteristics of C-MET mutations in patients with malignant melanoma. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4588. doi:1538-7445.AM2012-4588

Introduction: Ongoing clinical trials are based on targeting specific pathways in addition to the tumor histology. We examined the results of mutational analyses performed in patients with advanced cancer seen in the Phase I clinic at The University of Texas MD Anderson Cancer Center. Methods: Mutational analysis was mostly performed in the Clinical Laboratory Improvement Amendments (CLIA)-certified pathology laboratory of MD Anderson. DNA was extracted from microdissected paraffin-embedded tumor samples, and analysis was performed on specific exons, depending on the test ordered, for the following genes: PIK3CA (exon 9: codons 532-554; exon 20: codons 1011-1062); BRAF (exon 15: codons 595 to 600); KRAS and NRAS (codons 12, 13, and 61); EGFR (exons 18 to 21 of the kinase domain); KIT (exons 9, 11, 13, and 17); and RET (exon 10: codons 609, 611, 618, and 620; codon 634 of exon 11; codon 918 of exon 16). The loss of the tumor suppressor nuclear protein, PTEN, was determined using immunohistochemical staining. Results. Tumor mutational analysis was ordered for 952 patients. Overall, 103 patients did not have adequate tissue. Of the remaining 849 patients, 354 (41.69%) had ≥ 1 mutation and 495 did not have a mutation. More women (45%) than men (38%) had a mutation (p = 0.02), but age was not associated with the presence of mutations (p=0.76). Of the patients with mutations, 313 had 1 mutation, 38 had 2 mutations, and 3 had 3 mutations. The distribution of mutations by diagnosis is shown in the Table. The total distribution of mutations in our patient population was as follows: BRAF, 19.01%; KRAS, 18.74%; PIK3CA, 9.85%; NRAS, 8.17%; EGFR, 3.38%; KIT, 2.13%; and PTEN loss, 21.20%. Conclusion: Testing for PIK3CA, KRAS, NRAS, BRAF, EGFR, KIT, and RET mutations and PTEN loss in 849 patients with available tissue demonstrated that molecular driver aberrations are common in advanced cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4410. doi:10.1158/1538-7445.AM2011-4410

Proceedings: AACR 102nd Annual Meeting 2011‐‐ Apr 2‐6, 2011; Orlando, FL Introduction: To assess the outcomes of molecularly targeted therapy, we analyzed the results of mutational analyses in patients seen in our Phase I clinic and assessed patient survival by the number of mutations in patients treated with and without matched targeted therapy. Methods: Mutational analysis was mostly performed in the CLIA-certified pathology laboratory. DNA was extracted from microdissected paraffin-embedded tumor samples. The following genes were analyzed: PIK3CA (exon 9: codons 532-554; exon 20: codons 1011-1062); BRAF (exon 15: codons 595 to 600); KRAS and NRAS (codons 12, 13, 61); EGFR (exons 18 to 21, kinase domain); KIT (exons 9, 11, 13, 17); and RET (exon 10: codons 609, 611, 618, 620; exon 11: codon 634; exon 16: codon 918). The loss of the tumor suppressor nuclear protein (PTEN) was determined by immunohistochemistry. The allocation of patients to clinical trials varied according to protocol availability; eligibility; patient's prior response to therapy, toxicity and preference; and physician's choice. Patients whose tumor had a mutation were preferably treated with a matched targeted agent. Survival was compared between patients treated with and without matched targeted therapy. Results. Of 849 patients who had tissue available, 354 (41.7%) had ≥ 1 mutation (1, n=313; 2, n=38; 3, n=3) ([Table 1][1]). The median survival of patients with 0 vs. 1 vs. 2-3 mutations was 9.4 vs. 11.7 vs. 7.3 months, respectively (p =0.02; [Table 2][1]). In 278 evaluable patients with 1 mutation, the 2-yr survival rates were 42% and 17% (p = 0.02) for patients treated with and without matched targeted therapy. Conclusion: In patients with 1 mutation, treatment with matched targeted therapy was associated with superior survival compared to treatment without matching. Patients with 1 mutation had longer survival than those with no or >1 mutation, perhaps because most of them were treated with a matched targeted therapy. ![Figure][2] Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1287. doi:10.1158/1538-7445.AM2011-1287 [1]: #F1 [2]: pending:yes

Dasatinib is approximately 300 times more potent than imatinib in vitro and has significant activity in patients (pts) with CML-CP resistant or intolerant of imatinib (IM). We initiated a phase II trial to study the efficacy and safety of dasatinib in pts with previously untreated CML-CP. Aims: To investigate the efficacy and safety of dasatinib as initial therapy for patients with CML-CP. Methods: The primary objective was to estimate the proportion of pts attaining major molecular response at 12 months (mo). Pts with previously untreated CML-CP within 6 mo from diagnosis were eligible and received dasatinib 100 mg/day, randomized to either 50 mg twice daily (BID) or 100 mg once daily (QD). After 66 pts were accrued, the BID arm was closed and all subsequent pts were treated with 100 mg QD. Results: 82 pts have been enrolled (49 on the QD schedule, 33 BID). Median age was 47 years (yrs) (range 18-76 yrs). Median follow-up is 29 mo (range, 0 to 54 mo). All 72 pts who were not in CHR at the start of therapy achieved CHR. Among 75 pts followed for at least 3 mo, 73 (97%) achieved complete cytogenetic response (CCyR). Major molecular response (MMR) has been achieved in 64 (86%), including 19 (26%) with complete molecular response. The CCyR rate at different timepoints compares favorably to that observed in historical controls treated with imatinib 400mg or 800 mg daily. MMR was observed in 81% and 82% of the QD and BID patients respectively. However, the 12 mo MMR trended slightly higher with the BID dosing schedule compared to the QD schedule: 81% and 71% (p=0.384) respectively. Grade 3-4 non-hematologic toxicity included pain (muscle or joint) (13%), fatigue (11%), dyspnea (7%), neuropathy (6%), memory impairment, headache (5% each), diarrhea, pleural effusion, musculoskeletal complaints (2% each) and rash, prolonged QTC, weight gain and pruritus (1% each). Pleural effusion occurred in 16% evaluable pts. Grade 3-4 hematologic toxicity (transient) included thrombocytopenia 13%, neutropenia 25%, and anemia 9%. Fifty-two (63%) of 82 pts required transient treatment interruptions. The actual median daily dose for all pts was 100mg. There is no significant difference in grade 3-4 toxicity by treatment schedule but there was a trend for less pleural effusion with QD (13%) vs BID (22%; p=0.267). Five pts lost CCyR:(including 2 because of non-compliance). The 24-month probability of event-free survival (EFS) is 88%. There have been no transformations on study, and only one patient has died (metastatic pancreatic cancer). Ten patients have discontinued therapy (3 patient choice, 5 toxicity 2 pleural effusion, 1 prolonged thrombocytopenia, 1 bone pain, 1 congestive heart failure, and 2 for loss of MCyR) Conclusion: Rapid CCyR occurs in nearly all patients with previously untreated CML-CP treated with frontline dasatinib therapy; the MMR rate at 24 months was 93%, with a favorable toxicity profile. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4715. doi:10.1158/1538-7445.AM2011-4715