Steven J. Cohen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Steven J. Cohen has a total of 145 co-authors. Here are the top co-authors by number of shared publications:
- Igor Astsaturov
Fox Chase Cancer Center, Philadelphia, USA
7 shared publications Neal J. Meropol
University Hospitals Case Medical Center and Case Western Reserve University, Cleveland OH
7 shared publications- Joshua E. Meyer
Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, USA
6 shared publications Louis M. Weiner
Lombardi Cancer Center, Washington, DC
5 shared publicationsAndré Rogatko
Samuel Oschin Comprehensive Cancer Institute, Cedars-Sinai Medical Center, Los Angeles, CA
4 shared publications- Namrata Vijayvergia
Fox Chase Cancer Center, Philadelphia, PA
4 shared publications
Latest Publications
- 1 February 2017
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 35 (4_suppl) , 302
RNA profiling of neuroendocrine tumors (NETs) to define new targets and mechanisms of progression.
302 Background: NETs lack mutations in the “classical” signaling pathways but share mutations in regulators of gene expression (Jiao; 2011). We compared gene expression in PD & WD NETs to identify novel targets and biomarkers of differentiation. Methods: High quality RNA, extracted from paraffin blocks of deidentified NETs under an IRB-approved protocol, was profiled using a 770 gene panel (nCounter PanCancer pathway, Nanostring Technologies). The resulting data was used to identify the differentially expressed genes between PD and WD NETs using limma software (Ritchie; 2015). Gene Set Enrichment Analysis (Subramanian; 2005) identified differential pathway enrichment by calculating a Normalized Enrichment Score (NES). Results: Analysis of 16 PD and 23 WD NET samples identified 154 genes as extreme outliers ( > 2 fold up/downregulation between the subtypes). Compared to WD NETS, drug targets of interest overexpressed in PD NETs were histone lysine methyltransferase EZH2, and a cell cycle regulator CHEK1 (6.5x and 8.1x, respectively, p < 0.001). In contrast, serine/threonine protein kinase PAK 3 was upregulated in WD (10.6x, p < 0.001). These and other biomarkers will be further validated by immunolabeling of tissue sections. We also found differential enrichment of canonical pathways in PD versus WD NETs (table). Conclusions: Extreme outlier transcripts identified in PD & WD NETs support investigation of inhibitors of EZH2 (e.g. EPZ6438) and CHEK1 (e.g. LY2606368) in PD and PAK3(e.g. FRAX597) in WD NETs. Genes involved in cell cycle regulation and DNA repair in PD NETs and calcium / G protein coupled receptor signaling in WD NET account for biological differences between the 2 molecular subtypes and warrant future investigation as classifiers for NETs. Our findings provide mechanistic insights into the biology of NET and targets for therapy with direct clinical implications.[Table: see text]
- 1 February 2017
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 35 (4_suppl) , 302
RNA profiling of neuroendocrine tumors (NETs) to define new targets and mechanisms of progression.
302 Background: NETs lack mutations in the “classical” signaling pathways but share mutations in regulators of gene expression (Jiao; 2011). We compared gene expression in PD & WD NETs to identify novel targets and biomarkers of differentiation. Methods: High quality RNA, extracted from paraffin blocks of deidentified NETs under an IRB-approved protocol, was profiled using a 770 gene panel (nCounter PanCancer pathway, Nanostring Technologies). The resulting data was used to identify the differentially expressed genes between PD and WD NETs using limma software (Ritchie; 2015). Gene Set Enrichment Analysis (Subramanian; 2005) identified differential pathway enrichment by calculating a Normalized Enrichment Score (NES). Results: Analysis of 16 PD and 23 WD NET samples identified 154 genes as extreme outliers ( > 2 fold up/downregulation between the subtypes). Compared to WD NETS, drug targets of interest overexpressed in PD NETs were histone lysine methyltransferase EZH2, and a cell cycle regulator CHEK1 (6.5x and 8.1x, respectively, p < 0.001). In contrast, serine/threonine protein kinase PAK 3 was upregulated in WD (10.6x, p < 0.001). These and other biomarkers will be further validated by immunolabeling of tissue sections. We also found differential enrichment of canonical pathways in PD versus WD NETs (table). Conclusions: Extreme outlier transcripts identified in PD & WD NETs support investigation of inhibitors of EZH2 (e.g. EPZ6438) and CHEK1 (e.g. LY2606368) in PD and PAK3(e.g. FRAX597) in WD NETs. Genes involved in cell cycle regulation and DNA repair in PD NETs and calcium / G protein coupled receptor signaling in WD NET account for biological differences between the 2 molecular subtypes and warrant future investigation as classifiers for NETs. Our findings provide mechanistic insights into the biology of NET and targets for therapy with direct clinical implications.[Table: see text]
- 20 May 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (15_suppl) , 11604
Tumor profiling of liver metastases (LM) from CRC, NSCLC, pancreatic (PC), breast (BC) and gastroesophageal (GE) tumors to reveal differences versus primary tumors including in cMET, MYC, CDK4, Her2, b-catenin and PD1.
11604 Background: Liver is the most common metastatic site for various tumors and associates with poor prognosis. To understand LM as distinct molecularly from other origins such as CRC, NSCLC, BC and GE, we compared a large cohort of LM with local disease (LD) and other metastases (OM). Methods: Tumors from LD or metastatic site submitted to Caris Life Sciences for IHC (protein expression), ISH (gene amplification) and NGS sequencing (mutation, MT, CNV) between 2009 and 2015 were studied. A total of 9818 NSCLC (6289 LD, 420 LM, 3109 OM), 3486 PC (1453 LD, 1111 LM, 922 OM), 9739 BC (4413 LD, 1232 LM, 4094 OM), 1656 GE (1456 LD, 200 LM, 3937 OM) and 6368 CRC (4842 LD, 1526 LM, 3069 OM) were included. Tumors (metastases and primary) were not paired. Chi-square tests were used for statistics. Results: TOPO1 expression (Table) was highest in LM vs OM and LD, p < 0.01, suggesting a difference in DNA replication activity. Similar observations were made for TOP2A and TLE3. PD-L1 varied, and PD-1 on TILs was consistently the lowest in LM, and highest in LD (p < 0.02). Additionally, in CRC LM, OM and LD, Her2 ISH (5%, 2%, 3%), cMET IHC (56%, 51%, 48%) and APC (71%, 48%, 65%) MT were the highest in LM. In NSCLC, cMET ISH was higher in LM than LD (6.1% vs. 2.2%). MYC mutation was 11% in LM, 0.9% in OM and 0 in LD. In PC, cMET IHC was highest in LM (64%), followed by OM (57%) and LD (48%). In BC, CDK4 CNV was seen in 15% of LM, but not in OM or LD. In GE, CTNNB1 mutation was seen in 8% of LM, 2% in OM and 1% in LD. (p < 0.05). Conclusions: Molecular alterations shared by LM in 5 tumor types may warrant investigation in clinical trials of cMETi, BETi, CDKi, Her2, b-catenin and immune checkpoint inhibitors. Further research into alterations seen in LM may identify molecular drivers responsible for liver metastases. Differences between LM, OM, and LD support use of fresh biopsies from dominant metastatic site for molecular profiling.
- 20 May 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (15_suppl) , 11604
Tumor profiling of liver metastases (LM) from CRC, NSCLC, pancreatic (PC), breast (BC) and gastroesophageal (GE) tumors to reveal differences versus primary tumors including in cMET, MYC, CDK4, Her2, b-catenin and PD1.
11604 Background: Liver is the most common metastatic site for various tumors and associates with poor prognosis. To understand LM as distinct molecularly from other origins such as CRC, NSCLC, BC and GE, we compared a large cohort of LM with local disease (LD) and other metastases (OM). Methods: Tumors from LD or metastatic site submitted to Caris Life Sciences for IHC (protein expression), ISH (gene amplification) and NGS sequencing (mutation, MT, CNV) between 2009 and 2015 were studied. A total of 9818 NSCLC (6289 LD, 420 LM, 3109 OM), 3486 PC (1453 LD, 1111 LM, 922 OM), 9739 BC (4413 LD, 1232 LM, 4094 OM), 1656 GE (1456 LD, 200 LM, 3937 OM) and 6368 CRC (4842 LD, 1526 LM, 3069 OM) were included. Tumors (metastases and primary) were not paired. Chi-square tests were used for statistics. Results: TOPO1 expression (Table) was highest in LM vs OM and LD, p < 0.01, suggesting a difference in DNA replication activity. Similar observations were made for TOP2A and TLE3. PD-L1 varied, and PD-1 on TILs was consistently the lowest in LM, and highest in LD (p < 0.02). Additionally, in CRC LM, OM and LD, Her2 ISH (5%, 2%, 3%), cMET IHC (56%, 51%, 48%) and APC (71%, 48%, 65%) MT were the highest in LM. In NSCLC, cMET ISH was higher in LM than LD (6.1% vs. 2.2%). MYC mutation was 11% in LM, 0.9% in OM and 0 in LD. In PC, cMET IHC was highest in LM (64%), followed by OM (57%) and LD (48%). In BC, CDK4 CNV was seen in 15% of LM, but not in OM or LD. In GE, CTNNB1 mutation was seen in 8% of LM, 2% in OM and 1% in LD. (p < 0.05). Conclusions: Molecular alterations shared by LM in 5 tumor types may warrant investigation in clinical trials of cMETi, BETi, CDKi, Her2, b-catenin and immune checkpoint inhibitors. Further research into alterations seen in LM may identify molecular drivers responsible for liver metastases. Differences between LM, OM, and LD support use of fresh biopsies from dominant metastatic site for molecular profiling.
- 20 May 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (15_suppl) , e15158
Patients with rectal cancer after treatment of prostate cancer: A single institutition experience.
E15158 Background: Previous SEER studies has demonstrated that prostate cancer patients treated with radiation therapy have an increased risk of rectal cancer beginning approximately 5 years after prostate cancer treatment, but there are limited data on the treatment and outcomes of these patients. Objective: The aim of our study was to evaluate the presentation, treatment and survival of rectal cancer patients after prostate cancer treatment. Methods: This is a retrospective cohort study. We evaluated rectal cancer patients at a tertiary referral cancer center treated from 1990-2015 who had prostate cancer at least 6 months prior to diagnosis. Results: There were 53 rectal cancer patients diagnosed at a median age of 71 years (range 56-87). The median time to developing rectal cancer after prostate cancer was 5 years (range 0.5-16). Prostate cancer treatment included prostatectomy n = 14 (26.4%), radiation n = 36 (67.9%), hormonal therapy n = 2 (3.8%) or observation n = 1(1.9%). Thirty-two patients had clinical stage II/III rectal cancer, of these 6 (18.8%) received neoadjuvant chemoradiotherapy. Of the forty-six patients who underwent a surgical resection, 22 (41.5%) had an APR, 16 (30.2%) had a LAR and 8 (15.1%) had a local excision. An ostomy was required in 29 (63%) of these patients. Thirty patients received adjuvant therapy in the form of chemotherapy only 20 (66.7%), radiation therapy only 3 (10%) and combined therapies 7 (23.3%). Of note, five patients received neoadjuvant chemoradiation therapy for their rectal cancer after prostate radiation, with 3 out of 5 (60%) experiencing complications. The overall recurrence rate was 15/46 (32.6%) with a median time to recurrence of 1.84 years, (range 0.9-13.2). Six recurrences (40%) were initially local and 9 (60%) were initially distant. Conclusions: Patients who develop rectal cancer after prostate cancer present a challenging clinical scenario. In our small subset of patients there was a high morbidity associated with surgical resection following neoadjuvant chemoradiotherapy after prostate cancer radiotherapy. Future research is warranted to investigate the best sequencing of combined modality treatment to ensure the best outcomes in these complex patients.
- 20 May 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (15_suppl) , e15158
Patients with rectal cancer after treatment of prostate cancer: A single institutition experience.
E15158 Background: Previous SEER studies has demonstrated that prostate cancer patients treated with radiation therapy have an increased risk of rectal cancer beginning approximately 5 years after prostate cancer treatment, but there are limited data on the treatment and outcomes of these patients. Objective: The aim of our study was to evaluate the presentation, treatment and survival of rectal cancer patients after prostate cancer treatment. Methods: This is a retrospective cohort study. We evaluated rectal cancer patients at a tertiary referral cancer center treated from 1990-2015 who had prostate cancer at least 6 months prior to diagnosis. Results: There were 53 rectal cancer patients diagnosed at a median age of 71 years (range 56-87). The median time to developing rectal cancer after prostate cancer was 5 years (range 0.5-16). Prostate cancer treatment included prostatectomy n = 14 (26.4%), radiation n = 36 (67.9%), hormonal therapy n = 2 (3.8%) or observation n = 1(1.9%). Thirty-two patients had clinical stage II/III rectal cancer, of these 6 (18.8%) received neoadjuvant chemoradiotherapy. Of the forty-six patients who underwent a surgical resection, 22 (41.5%) had an APR, 16 (30.2%) had a LAR and 8 (15.1%) had a local excision. An ostomy was required in 29 (63%) of these patients. Thirty patients received adjuvant therapy in the form of chemotherapy only 20 (66.7%), radiation therapy only 3 (10%) and combined therapies 7 (23.3%). Of note, five patients received neoadjuvant chemoradiation therapy for their rectal cancer after prostate radiation, with 3 out of 5 (60%) experiencing complications. The overall recurrence rate was 15/46 (32.6%) with a median time to recurrence of 1.84 years, (range 0.9-13.2). Six recurrences (40%) were initially local and 9 (60%) were initially distant. Conclusions: Patients who develop rectal cancer after prostate cancer present a challenging clinical scenario. In our small subset of patients there was a high morbidity associated with surgical resection following neoadjuvant chemoradiotherapy after prostate cancer radiotherapy. Future research is warranted to investigate the best sequencing of combined modality treatment to ensure the best outcomes in these complex patients.
- 1 February 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (4_suppl) , 415
The association between sarcopenia and treatment outcomes in patients with pancreatic cancer undergoing chemoradiation.
415 Background: Sarcopenia (Sp) or severe loss of muscle mass is a prognostic factor in cancer patients. We assessed the association between Sp and outcomes in borderline resectable or locally advanced/unresectable pancreatic cancer patients undergoing chemoradiation (CRT). Methods: We reviewed all patients who underwent CRT at an NCI-designated cancer center between 2007-2012. Patients with available pre-treatment imaging were included in the analysis. Sp was defined as a lumbar skeletal muscle area/height of < 55.4 cm2/m2 for males and < 38.9 cm2/m2for females. Sp was correlated to treatment toxicity, cause specific survival (CSS), and overall survival (OS). Analysis was performed using chi-square test for categorical variables and Mann-Whitney U test for continuous variables. Kaplan-Meier method and Cox regression was used for survival analysis. Results: A total of 86 patients met the inclusion criteria with 32 (37%) patients being sarcopenic. The median age was 70 (range 46-91) with a median follow-up of 14 months (3-68). The majority of patients were male (57%), had T3 disease (47%), or were borderline resectable (63%). Twenty nine (33.7%) patients underwent surgery. Sarcopenic patients were less likely to undergo surgery (p = 0.024) versus non-sarcopenic patients. Otherwise there was no difference in clinical or treatment factors. The 12-month actuarial OS for patients with and without Sp was 47.9% and 70.7%, respectively (p = 0.047). The 12-month actuarial CSS for patients with and without Sp was 48% and 76%, respectively (p = 0.007). On multivariable analysis, after controlling for T-stage, N-stage, resectability, gender, pre-treatment CA 19-9, and surgery, Sp was no longer associated with CSS (HR 0.284 95% CI 0.774-2.398) or OS (HR 1.077 95% CI 0.626-1.853). Surgery remained associated with CSS (HR 0.205 95% CI = 0.102-0.412) and OS (HR 0.187 95% CI 0.096-0.363). T-stage also remained associated with both CSS (HR 0.616 95% CI 0.410-0.925) and OS (HR 0.649 95% CI 0.440-0.957). Conclusions: The presence of Sp decreases the likelihood of undergoing curative surgery for pancreatic cancer. This may be another useful tool to identify poor prognosis patients.
- 1 February 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (4_suppl) , 415
The association between sarcopenia and treatment outcomes in patients with pancreatic cancer undergoing chemoradiation.
415 Background: Sarcopenia (Sp) or severe loss of muscle mass is a prognostic factor in cancer patients. We assessed the association between Sp and outcomes in borderline resectable or locally advanced/unresectable pancreatic cancer patients undergoing chemoradiation (CRT). Methods: We reviewed all patients who underwent CRT at an NCI-designated cancer center between 2007-2012. Patients with available pre-treatment imaging were included in the analysis. Sp was defined as a lumbar skeletal muscle area/height of < 55.4 cm2/m2 for males and < 38.9 cm2/m2for females. Sp was correlated to treatment toxicity, cause specific survival (CSS), and overall survival (OS). Analysis was performed using chi-square test for categorical variables and Mann-Whitney U test for continuous variables. Kaplan-Meier method and Cox regression was used for survival analysis. Results: A total of 86 patients met the inclusion criteria with 32 (37%) patients being sarcopenic. The median age was 70 (range 46-91) with a median follow-up of 14 months (3-68). The majority of patients were male (57%), had T3 disease (47%), or were borderline resectable (63%). Twenty nine (33.7%) patients underwent surgery. Sarcopenic patients were less likely to undergo surgery (p = 0.024) versus non-sarcopenic patients. Otherwise there was no difference in clinical or treatment factors. The 12-month actuarial OS for patients with and without Sp was 47.9% and 70.7%, respectively (p = 0.047). The 12-month actuarial CSS for patients with and without Sp was 48% and 76%, respectively (p = 0.007). On multivariable analysis, after controlling for T-stage, N-stage, resectability, gender, pre-treatment CA 19-9, and surgery, Sp was no longer associated with CSS (HR 0.284 95% CI 0.774-2.398) or OS (HR 1.077 95% CI 0.626-1.853). Surgery remained associated with CSS (HR 0.205 95% CI = 0.102-0.412) and OS (HR 0.187 95% CI 0.096-0.363). T-stage also remained associated with both CSS (HR 0.616 95% CI 0.410-0.925) and OS (HR 0.649 95% CI 0.440-0.957). Conclusions: The presence of Sp decreases the likelihood of undergoing curative surgery for pancreatic cancer. This may be another useful tool to identify poor prognosis patients.
- 1 February 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (4_suppl) , 386
Nutritional status and outcomes of older versus younger patients (pts) with metastatic pancreatic cancer (mPC).
386 Background: Older mPC pts are less likely to receive chemotherapy (Ctx) and receive fewer agents compared to younger pts, yet have similar overall survival (OS). Both obesity and malnutrition are linked with poorer outcomes in pancreatic cancer. We evaluated the effect of body mass index (BMI) and nutritional state on survival in < 65 y (GpA) and > 65 y (GpB) mPC pts. Methods: With IRB approval, we retrospectively analyzed charts of 579 mPC pts treated between 2000 and 2010. GpA and GpB were compared based on BMI alone (25) and also their nutritional status (poor: albumin 15% during disease course with any BMI; good: Alb>3.5 + Wt loss20; intermediate: all others). Log-rank tests and Cox proportional hazards models were used to analyze OS and Fisher’s exact test to compare categorical variables. Results: There were 299 pts in GpA (median age 57yr) and 280 pts in GpB (median age 73 yr). Information on BMI/albumin was missing for 126 pts; total study population was 454 (GpA 238, GpB 216).Groups were well balanced for gender (majority male), PS (majority ≤1), initial stage (majority IV), and primary site (majority head). There was no significant difference in Ctx and number of agents used between the 3 subgroups across GpA and GpB. BMI alone did not affect survival in both groups. Nutritional status was significantly associated with survival when evaluating the whole population and GpA, but not GpB (Table). Good nutrition GpA pts had a lower relative risk of death versus pts in intermediate and poor nutrition groups (RR 0.83, CI 0.71-0.96, p=0.02). Good nutritional status was associated with better survival in GpA on multivariate analysis (RR=0.81, CI 0.69-0.95; p=0.008,), but not in GpB. Conclusions: Nutritional status (defined by BMI, albumin and weight loss) was associated with survival in younger mPC pts but not in older pts. Greater attention to nutrition and cancer cachexia with possible interventions throughout treatment may improve survival in mPC pts undergoing chemotherapy, especially those < 65 yrs. Table [Table: see text]
- 1 February 2016
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 34 (4_suppl) , 386
Nutritional status and outcomes of older versus younger patients (pts) with metastatic pancreatic cancer (mPC).
386 Background: Older mPC pts are less likely to receive chemotherapy (Ctx) and receive fewer agents compared to younger pts, yet have similar overall survival (OS). Both obesity and malnutrition are linked with poorer outcomes in pancreatic cancer. We evaluated the effect of body mass index (BMI) and nutritional state on survival in < 65 y (GpA) and > 65 y (GpB) mPC pts. Methods: With IRB approval, we retrospectively analyzed charts of 579 mPC pts treated between 2000 and 2010. GpA and GpB were compared based on BMI alone (25) and also their nutritional status (poor: albumin 15% during disease course with any BMI; good: Alb>3.5 + Wt loss20; intermediate: all others). Log-rank tests and Cox proportional hazards models were used to analyze OS and Fisher’s exact test to compare categorical variables. Results: There were 299 pts in GpA (median age 57yr) and 280 pts in GpB (median age 73 yr). Information on BMI/albumin was missing for 126 pts; total study population was 454 (GpA 238, GpB 216).Groups were well balanced for gender (majority male), PS (majority ≤1), initial stage (majority IV), and primary site (majority head). There was no significant difference in Ctx and number of agents used between the 3 subgroups across GpA and GpB. BMI alone did not affect survival in both groups. Nutritional status was significantly associated with survival when evaluating the whole population and GpA, but not GpB (Table). Good nutrition GpA pts had a lower relative risk of death versus pts in intermediate and poor nutrition groups (RR 0.83, CI 0.71-0.96, p=0.02). Good nutritional status was associated with better survival in GpA on multivariate analysis (RR=0.81, CI 0.69-0.95; p=0.008,), but not in GpB. Conclusions: Nutritional status (defined by BMI, albumin and weight loss) was associated with survival in younger mPC pts but not in older pts. Greater attention to nutrition and cancer cachexia with possible interventions throughout treatment may improve survival in mPC pts undergoing chemotherapy, especially those < 65 yrs. Table [Table: see text]
- 20 January 2015
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 33 (3_suppl) , 407
Neoadjuvant chemoradiation (NACRT) and the prognostic effect of surgical margin (SM) status in pancreatic adenocarcinoma (PAC).
407 Background: Many studies have associated a R0 resection to have significantly improved survival compared with a R1 resection in PAC. Patients (pts) who undergo NACRT often go to surgery 4-8 weeks after the end of therapy, before the effects of NACRT can be fully manifested. The goal of this study is to evaluate if a positive SM (+SM) after NACRT has the same poor prognosis as a +SM after upfront surgery. Methods: After IRB approval, we retrospectively reviewed all cases of surgically resected PAC at a single institution from Dec 1996 to Jan 2014. Pts were stratified by receipt of NACRT as well as by SM status. We excluded pts treated with palliative intent, metastatic disease at presentation, death within 90 days of surgery, and biliary or ampullary tumors. The primary endpoint was overall survival (OS). We assessed the relationship between pt and tumor variables with treatment/margin combination using Chi-squared tests. OS was examined using Kaplan-Meier curves, and we tested association with treatment/margin using log-rank tests. Results: A total of 213 pts met inclusion criteria; 111 received upfront surgery (group I) with 94 (85%) receiving adjuvant chemotherapy or CRT and 102 received NACRT (group II) with either concurrent 5-fluorouracil (n=18) or gemcitabine (n=84). There were 31 pts with +SM in group I and 29 pts in group II. Pt demographics were balanced. There was more vessel involvement in group II (81%) at diagnosis than group I (11%) (p
- 20 January 2015
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 33 (3_suppl) , 407
Neoadjuvant chemoradiation (NACRT) and the prognostic effect of surgical margin (SM) status in pancreatic adenocarcinoma (PAC).
407 Background: Many studies have associated a R0 resection to have significantly improved survival compared with a R1 resection in PAC. Patients (pts) who undergo NACRT often go to surgery 4-8 weeks after the end of therapy, before the effects of NACRT can be fully manifested. The goal of this study is to evaluate if a positive SM (+SM) after NACRT has the same poor prognosis as a +SM after upfront surgery. Methods: After IRB approval, we retrospectively reviewed all cases of surgically resected PAC at a single institution from Dec 1996 to Jan 2014. Pts were stratified by receipt of NACRT as well as by SM status. We excluded pts treated with palliative intent, metastatic disease at presentation, death within 90 days of surgery, and biliary or ampullary tumors. The primary endpoint was overall survival (OS). We assessed the relationship between pt and tumor variables with treatment/margin combination using Chi-squared tests. OS was examined using Kaplan-Meier curves, and we tested association with treatment/margin using log-rank tests. Results: A total of 213 pts met inclusion criteria; 111 received upfront surgery (group I) with 94 (85%) receiving adjuvant chemotherapy or CRT and 102 received NACRT (group II) with either concurrent 5-fluorouracil (n=18) or gemcitabine (n=84). There were 31 pts with +SM in group I and 29 pts in group II. Pt demographics were balanced. There was more vessel involvement in group II (81%) at diagnosis than group I (11%) (p
- 20 January 2015
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 33 (3_suppl) , 245
Molecular profiling of neuroendocrine tumors (NETs): The Fox Chase Cancer Center (FCCC) experience.
245 Background: The rarity of NETs can limit clinical trial accrual to develop new therapies. Given fewer approved treatments, a better understanding of underlying biology is critical to development of and assignment of patients (pts) to clinical trials. Methods: Pts with NETs (excluding small/large cell lung cancer) of all grades at FCCC were enrolled onto a prospective IRB approved protocol that utilized an NGS platform to detect somatic mutations (SM) in 50 cancer-related genes (Cancer Code) on archived tissue from primary or metastatic sites. Genes tested included ABL1, AKT1, ALK, APC, ATM, BRAF, CDH1, CDKN2A, CSF1R, CTNNB1, EGFR, ERBB2, ERBB4, EZH2, FBXW7, FGFR1, FGFR2, FGFR3, FLT3, GNA11, GNAQ, GNAS, HNF1A, HRAS, IDH1, IDH2, JAK2, JAK3, KDR, KIT, KRAS, MET, MLH1, MPL, NOTCH1, NPM1, NRAS, PDGFRA, PIK3CA, PTEN, PTPN11, RB1, RET, SMAD4, SMARCB1, SMO,SRC, STK11, TP53 and VHL. Review of pathology specimens for grade and Ki -67 was also performed. Results: Thirty-nine pts (median age 59 y, males 46%) were enrolled from October 2013 to July 2014. Gene profiling results are available on thirty-five pts. Ki-67 score was reviewed for 31/35 tumors. 6 (20%) pts had high grade (HG) tumors (Ki-67 > 20%) and 25 (80%) had low/intermediate grade (LIG) tumors (Ki-67 ≤20%). Thirteen (37%) pts were found to have SMs and 22 (63%) did not, with 4 (12%) pts’ tumors having >1 SMs (2 HG and 2 LIG tumors). Incidence of SM was 41% (12/29) in Caucasians, 16% (1/6) in other races, 46% (6/13) in smokers and 30% (6/20) in non-smokers. Incidence of SM was 24% (6/25) in LIG NETs and 84% (5/6) in the HG NETs. Among HG tumors, 66% (4/6) harbored TP53 gene mutation and 33% (2/6) were BRAF mutation positive. Conclusions: Tumor-specific mutations are seen in a minority of low grade NETs but are common in high grade tumors. Interestingly, no mutations were identified in pts with unknown primary. Analysis of clinical outcomes based on treatment received is ongoing to assess for possible prognostic/therapeutic implications of these mutations. [Table: see text]
- 20 January 2015
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 33 (3_suppl) , 245
Molecular profiling of neuroendocrine tumors (NETs): The Fox Chase Cancer Center (FCCC) experience.
245 Background: The rarity of NETs can limit clinical trial accrual to develop new therapies. Given fewer approved treatments, a better understanding of underlying biology is critical to development of and assignment of patients (pts) to clinical trials. Methods: Pts with NETs (excluding small/large cell lung cancer) of all grades at FCCC were enrolled onto a prospective IRB approved protocol that utilized an NGS platform to detect somatic mutations (SM) in 50 cancer-related genes (Cancer Code) on archived tissue from primary or metastatic sites. Genes tested included ABL1, AKT1, ALK, APC, ATM, BRAF, CDH1, CDKN2A, CSF1R, CTNNB1, EGFR, ERBB2, ERBB4, EZH2, FBXW7, FGFR1, FGFR2, FGFR3, FLT3, GNA11, GNAQ, GNAS, HNF1A, HRAS, IDH1, IDH2, JAK2, JAK3, KDR, KIT, KRAS, MET, MLH1, MPL, NOTCH1, NPM1, NRAS, PDGFRA, PIK3CA, PTEN, PTPN11, RB1, RET, SMAD4, SMARCB1, SMO,SRC, STK11, TP53 and VHL. Review of pathology specimens for grade and Ki -67 was also performed. Results: Thirty-nine pts (median age 59 y, males 46%) were enrolled from October 2013 to July 2014. Gene profiling results are available on thirty-five pts. Ki-67 score was reviewed for 31/35 tumors. 6 (20%) pts had high grade (HG) tumors (Ki-67 > 20%) and 25 (80%) had low/intermediate grade (LIG) tumors (Ki-67 ≤20%). Thirteen (37%) pts were found to have SMs and 22 (63%) did not, with 4 (12%) pts’ tumors having >1 SMs (2 HG and 2 LIG tumors). Incidence of SM was 41% (12/29) in Caucasians, 16% (1/6) in other races, 46% (6/13) in smokers and 30% (6/20) in non-smokers. Incidence of SM was 24% (6/25) in LIG NETs and 84% (5/6) in the HG NETs. Among HG tumors, 66% (4/6) harbored TP53 gene mutation and 33% (2/6) were BRAF mutation positive. Conclusions: Tumor-specific mutations are seen in a minority of low grade NETs but are common in high grade tumors. Interestingly, no mutations were identified in pts with unknown primary. Analysis of clinical outcomes based on treatment received is ongoing to assess for possible prognostic/therapeutic implications of these mutations. [Table: see text]
- 20 January 2014
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 32 (3) , 214
Profiling of a global cohort of 1,250 neuroendocrine tumors to identify multiple potential drug targets.
214 Background: Identification of new drug targets may extend treatment options for NET, regardless of histologic classification or primary organ site. Methods: 1,250 cases of infradiaphragmatic neuroendocrine tumors (all grades and sites) were identified among >50,000 cases profiled in a CLIA certified laboratory. Biomarker profiling utilized multiple platforms: gene sequencing (next generation sequencing, Sanger or pyrosequencing), gene copy number by in-situ hybridization, and protein expression by immunohistochemistry. The results are shown relative to the total number of tests performed. Results: Overall, drug therapy-relevant alterations were identified in 1,130 of 1,250 (90%) of cases. Low or absent (0 or 1+ by IHC) MGMT, a biomarker of sensitivity to alkylating agents, was found in 130/219 pancreatic cases (59%), and in 450/991 (45%) of non-pancreatic NET. Low or absent (0 or 1+ by IHC) expression of RRM1, a biomarker of gemcitabine sensitivity, was found in 813/1,100 of NET (74%) and low or absent thymidine synthase, TS, a biomarker of fluoropyrimidine sensitivity, was shown for 793/1,096 (72%) of NET by IHC. Sequencing of tumors showed oncogenic mutations in BRAF (4/369 (V600E in 3 and G596R in 1), CTNNB1 (2/150), KIT (3/281), EGFR (1/178), FGFR2 (1/150), GNAS (1/150), HRAS (2/150), PIK3CA (6/343), RB (2/150) VHL (1/150), KRAS (10/125), NRAS (2/274), and APC (2/150) and amplifications of EGFR (46/686) and MET (4/236). Ki67 status and correlation between the site of origin and biomarkers will be presented. Therapies guided by mechanism-based biomarkers produced durable responses in documented cases: partial response (PR) >1 year to imatinib in a patient with KIT-mutant metastatic NET, and in cases of MGMTlow/TSlowtreated with streptozocine or temizolomide plus fluoropyrimidine chemotherapy, thus supporting the clinical relevance of target profiling in NET. Conclusions: Comprehensive multiplatform profiling of a large series (n=1,250) of NET, despite low frequency of individual biomarkers, identified clinically relevant targets in the majority of patients. Our results provide the basis for future clinical trials to assess the efficacy of biomarker-based therapy for NET.
- 20 January 2014
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 32 (3) , 214
Profiling of a global cohort of 1,250 neuroendocrine tumors to identify multiple potential drug targets.
214 Background: Identification of new drug targets may extend treatment options for NET, regardless of histologic classification or primary organ site. Methods: 1,250 cases of infradiaphragmatic neuroendocrine tumors (all grades and sites) were identified among >50,000 cases profiled in a CLIA certified laboratory. Biomarker profiling utilized multiple platforms: gene sequencing (next generation sequencing, Sanger or pyrosequencing), gene copy number by in-situ hybridization, and protein expression by immunohistochemistry. The results are shown relative to the total number of tests performed. Results: Overall, drug therapy-relevant alterations were identified in 1,130 of 1,250 (90%) of cases. Low or absent (0 or 1+ by IHC) MGMT, a biomarker of sensitivity to alkylating agents, was found in 130/219 pancreatic cases (59%), and in 450/991 (45%) of non-pancreatic NET. Low or absent (0 or 1+ by IHC) expression of RRM1, a biomarker of gemcitabine sensitivity, was found in 813/1,100 of NET (74%) and low or absent thymidine synthase, TS, a biomarker of fluoropyrimidine sensitivity, was shown for 793/1,096 (72%) of NET by IHC. Sequencing of tumors showed oncogenic mutations in BRAF (4/369 (V600E in 3 and G596R in 1), CTNNB1 (2/150), KIT (3/281), EGFR (1/178), FGFR2 (1/150), GNAS (1/150), HRAS (2/150), PIK3CA (6/343), RB (2/150) VHL (1/150), KRAS (10/125), NRAS (2/274), and APC (2/150) and amplifications of EGFR (46/686) and MET (4/236). Ki67 status and correlation between the site of origin and biomarkers will be presented. Therapies guided by mechanism-based biomarkers produced durable responses in documented cases: partial response (PR) >1 year to imatinib in a patient with KIT-mutant metastatic NET, and in cases of MGMTlow/TSlowtreated with streptozocine or temizolomide plus fluoropyrimidine chemotherapy, thus supporting the clinical relevance of target profiling in NET. Conclusions: Comprehensive multiplatform profiling of a large series (n=1,250) of NET, despite low frequency of individual biomarkers, identified clinically relevant targets in the majority of patients. Our results provide the basis for future clinical trials to assess the efficacy of biomarker-based therapy for NET.
- 20 January 2014
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 32 (3) , 126
Freedom from recurrence after induction cisplatin/5-FU/RT versus carboplatin/paclitaxel/RT in patients with esophageal cancer.
126 Background: The optimal neoadjuvant treatment regimen for esophageal cancer has yet to be defined. Methods: We analyzed a prospective database of esophageal cancer patients treated with cisplatin/5-FU/RT (A) or carboplatin/paclitaxel/RT (B) then esophagectomy. Endpoints were pathologic response (pCR), recurrence, and overall survival. Covariates (age, gender, clinical stage, Charlson comorbidity index (CCI), and ECOG PS) were adjusted via propensity score weighting from logistic regression. Differences in survival curves were compared using weighted Kaplan Meier methods and Wilcoxon tests. Landmark analysis compared differences in outcomes at least six months after surgery. Results: From January 2008 to June 2013, 71 patients received induction RT (50.4 Gy), with 38 and 33 patients in groups A and B, respectively. Patient demographics were similar (Table 1). Operative procedure, R0 resections, and operative complications did not differ (operative mortality 1/71, 1.4%; anastomotic leak rate 10% A vs. 9% B, p=1.0). Median follow-up was 18.2 and 9.1 months for groups A and B, respectively. There was no difference in overall survival (p=0.88) or total number of recurrences (11/38, 29% A vs. 10/33, 30% B). Landmark analysis revealed a difference in recurrence at least 6 months after surgery: time to recurrence was shorter in group B (median time to recurrence 31 mos A vs. 18 mos B, p=0.02). pCR was seen in 13/38 (34%) A and 9/33 (27%) B (p=0.53). In patients with pCR, there was a trend toward more recurrences in group B (2/13, 15% A v 4/9, 44% B, p=0.14). Conclusions: Six months after surgery patients receiving induction carboplatin/paclitaxel/RT had shorter time to recurrence compared to cisplatin/5-FU/RT. In those with pCR a trend towards increased distant recurrences was seen with carboplatin/paclitaxel/RT, which may have less systemic efficacy in esophageal cancer than cisplatin/5-FU/RT. This should be evaluated in larger series. [Table: see text]
- 20 January 2014
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 32 (3) , 126
Freedom from recurrence after induction cisplatin/5-FU/RT versus carboplatin/paclitaxel/RT in patients with esophageal cancer.
126 Background: The optimal neoadjuvant treatment regimen for esophageal cancer has yet to be defined. Methods: We analyzed a prospective database of esophageal cancer patients treated with cisplatin/5-FU/RT (A) or carboplatin/paclitaxel/RT (B) then esophagectomy. Endpoints were pathologic response (pCR), recurrence, and overall survival. Covariates (age, gender, clinical stage, Charlson comorbidity index (CCI), and ECOG PS) were adjusted via propensity score weighting from logistic regression. Differences in survival curves were compared using weighted Kaplan Meier methods and Wilcoxon tests. Landmark analysis compared differences in outcomes at least six months after surgery. Results: From January 2008 to June 2013, 71 patients received induction RT (50.4 Gy), with 38 and 33 patients in groups A and B, respectively. Patient demographics were similar (Table 1). Operative procedure, R0 resections, and operative complications did not differ (operative mortality 1/71, 1.4%; anastomotic leak rate 10% A vs. 9% B, p=1.0). Median follow-up was 18.2 and 9.1 months for groups A and B, respectively. There was no difference in overall survival (p=0.88) or total number of recurrences (11/38, 29% A vs. 10/33, 30% B). Landmark analysis revealed a difference in recurrence at least 6 months after surgery: time to recurrence was shorter in group B (median time to recurrence 31 mos A vs. 18 mos B, p=0.02). pCR was seen in 13/38 (34%) A and 9/33 (27%) B (p=0.53). In patients with pCR, there was a trend toward more recurrences in group B (2/13, 15% A v 4/9, 44% B, p=0.14). Conclusions: Six months after surgery patients receiving induction carboplatin/paclitaxel/RT had shorter time to recurrence compared to cisplatin/5-FU/RT. In those with pCR a trend towards increased distant recurrences was seen with carboplatin/paclitaxel/RT, which may have less systemic efficacy in esophageal cancer than cisplatin/5-FU/RT. This should be evaluated in larger series. [Table: see text]
- 1 February 2012
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 30 (4) , 74
A phase I evaluation of vandetanib plus paclitaxel, carboplatin, 5-fluorouracil, and XRT induction therapy followed by surgery for previously untreated locally advanced cancer of the esophagus and GE junction.
74 Background: Vandetanib (Vb) is an oral inhibitor that selectively targets VEGFR2, VEGFR3, RET and EGFR. Preclinical models indicate synergy with radiation and chemotherapy. We report the final results from an open-label phase I study of pre-operative Vb with carboplatin, paclitaxel, 5FU and radiation in pts with operable esophageal and GE junction (GEJ) carcinomas. Methods: Pts with operable PET/CT and EUS-staged carcinoma of the esophagus and GEJ received once daily Vb with concomitant RT (1.8 Gy/d, total 45 Gy) plus carboplatin AUC5 (d1 and 29), paclitaxel (50 mg/m2, d1,8,15,22,29), and 5-FU (CI 225 mg/m2, M-F, 5 wks). The primary objective was the safety, tolerability and the maximum tolerated dose of Vb in this combination. Planned dose escalations of Vb: 100, 200 and 300 mg daily with a “3+3” design. Results: Nine male pts (median age 58 (35-77), stage II, T2-3N0 (2 pts), III, T2-3N+ (7 pts)) were accrued between 03/2009 and 08/2010 at Fox Chase Cancer Center: 2/9 pts had esophageal squamous carcinoma (SCC) and 7/9 pts had adenocarcinoma including GEJ (2 pts). All patients completed the planned induction chemoradiation and underwent esophagectomy with curative intent. Of 3 initial pts at Vb 100 mg, one with SCC died on day 99 from GI hemorrhage which was considered treatment-related. Three more pts received Vb 100 mg and 3 pts received Vb 200 mg daily. Dose limiting toxicities (DLT) occurred in 2/3 pts receiving Vb 200 mg (esophago-gastric anastomotic leak, and diarrhea requiring Vb dose reduction). Grade 3 non-hematological toxicities included diarrhea (2), anorexia (1), abdominal pain (1), hyponatremia (1), AST/ALT (1), depression (1). Dose escalation to 300 mg Vb was not attempted. 5/9 pts had a pathologic complete response (no tumor seen, 56%), 5/9 remain disease free with median followup of 23 months (13-31). 3 pts had distant recurrences at 322, 334 and 561 days. Conclusions: Targeting VEGFR/RET/EGFR with Vb at 100 mg daily in combination with induction chemoradiotherapy for operable esophageal and GEJ cancer is well tolerated and with promising clinical activity warranting further phase II evaluation.
- 1 February 2012
- Published by American Society of Clinical Oncology (ASCO) in Journal of Clinical Oncology
- Vol. 30 (4) , 74
A phase I evaluation of vandetanib plus paclitaxel, carboplatin, 5-fluorouracil, and XRT induction therapy followed by surgery for previously untreated locally advanced cancer of the esophagus and GE junction.
74 Background: Vandetanib (Vb) is an oral inhibitor that selectively targets VEGFR2, VEGFR3, RET and EGFR. Preclinical models indicate synergy with radiation and chemotherapy. We report the final results from an open-label phase I study of pre-operative Vb with carboplatin, paclitaxel, 5FU and radiation in pts with operable esophageal and GE junction (GEJ) carcinomas. Methods: Pts with operable PET/CT and EUS-staged carcinoma of the esophagus and GEJ received once daily Vb with concomitant RT (1.8 Gy/d, total 45 Gy) plus carboplatin AUC5 (d1 and 29), paclitaxel (50 mg/m2, d1,8,15,22,29), and 5-FU (CI 225 mg/m2, M-F, 5 wks). The primary objective was the safety, tolerability and the maximum tolerated dose of Vb in this combination. Planned dose escalations of Vb: 100, 200 and 300 mg daily with a “3+3” design. Results: Nine male pts (median age 58 (35-77), stage II, T2-3N0 (2 pts), III, T2-3N+ (7 pts)) were accrued between 03/2009 and 08/2010 at Fox Chase Cancer Center: 2/9 pts had esophageal squamous carcinoma (SCC) and 7/9 pts had adenocarcinoma including GEJ (2 pts). All patients completed the planned induction chemoradiation and underwent esophagectomy with curative intent. Of 3 initial pts at Vb 100 mg, one with SCC died on day 99 from GI hemorrhage which was considered treatment-related. Three more pts received Vb 100 mg and 3 pts received Vb 200 mg daily. Dose limiting toxicities (DLT) occurred in 2/3 pts receiving Vb 200 mg (esophago-gastric anastomotic leak, and diarrhea requiring Vb dose reduction). Grade 3 non-hematological toxicities included diarrhea (2), anorexia (1), abdominal pain (1), hyponatremia (1), AST/ALT (1), depression (1). Dose escalation to 300 mg Vb was not attempted. 5/9 pts had a pathologic complete response (no tumor seen, 56%), 5/9 remain disease free with median followup of 23 months (13-31). 3 pts had distant recurrences at 322, 334 and 561 days. Conclusions: Targeting VEGFR/RET/EGFR with Vb at 100 mg daily in combination with induction chemoradiotherapy for operable esophageal and GEJ cancer is well tolerated and with promising clinical activity warranting further phase II evaluation.