5/1/15 1 Department of Medicine 5.1.15 Emily K. Bergsland, MD UCSF Helen Diller Family Comprehensive Cancer Center MANAGEMENT OF PANCREATIC NEUROENDOCRINE TUMORS DISCLOSURES: Support for clinical trials o Genentech (drug only) o Novartis o Lexicon Advisory Board (uncompensated) o Novartis o Celgene o Lexicon 2
37
Embed
MANAGEMENT OF PANCREATIC NEUROENDOCRINE TUMORS › 2015 › MDM15E01 › slides › 14... · among patients with progressive advanced pancreatic neuroendocrine tumors as compared
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
5/1/15
1
Department of Medicine
5.1.15
Emily K. Bergsland, MD
UCSF Helen Diller Family Comprehensive Cancer Center
MANAGEMENT OF PANCREATIC
NEUROENDOCRINE TUMORS
DISCLOSURES:
§ Support for clinical trials o Genentech (drug only) o Novartis o Lexicon
§ Advisory Board (uncompensated) o Novartis o Celgene o Lexicon
2
5/1/15
2
Tumors arising from enterochromaffin cells located in neuroendocrine tissue throughout the body
NETs can be functional or nonfunctional and include a heterogeneous group of neoplasms1,2
NEUROENDOCRINE TUMORS (NET): A DIVERSE GROUP OF MALIGNANCIES
References: 1. Dorland’s Medical Dictionary Web site. Available at: http://www.dorlands.com. Accessed November 10, 2008. 2. Modlin IM, Kidd M, Latich I,2. Zikusoka MN, Shapiro MD. Current status of gastrointestinal carcinoids. Gastroenterology. 2005;128:1717-1751.
§ Functional imaging for selection of patients who may benefit from receptor-based therapies
7 deHerder, 2014
GA68 DOTATOC PET-CT (INVESTIGATIONAL):
Negative study Liver met and pancreatic 1°
Courtesy of Thomas Hope, MD (UCSF)
5/1/15
5
GA68 DOTATOC PET-CT (INVESTIGATIONAL):
Courtesy of Thomas Hope, MD (UCSF)
NEUROENDOCRINE (NE) TUMORS OF THE GI TRACT: CLASSIFICATION
• Functioning: clinical syndrome due to hormone (<30%)
• serotonin, substance P, gastrin, insulin, glucagon, somatostatin, vasoactive intestinal peptide, growth hormone-releasing factor, adrenocorticotropic hormone • Multiple hormones posible, as is conversion from nonfunctional to functional over time
• Nonfunctioning: no specific symptoms • may also secrete hormones (pancreatic polypeptide)
5/1/15
6
References: 1. Strosberg JR, Nasir A, Hodul P, Kvols L. GI Cancer Res. 2008; 2:113-125. 2. Klöppel G, Perren A, Heitz PU. Ann Ny Acad Sci. 2004; 1014:13-27.
Good Poor
SPECTRUM OF NEUROENDOCRINE TUMORS: PROGNOSIS
NET CLASSIFICATION 186 Chen et al. Molecular pathology of pancreatic neuroendocrine tumors
PETs (Table 2� ���� � *O� UIF�8)0������DMBTTJGJDBUJPO �the malignant potential of pancreatic neuroendocrine neoplasms is acknowledged and enforced. The fact is that PETs are often malignant because they are metastatic at diagnosis, or at least have the potential to metastasize in a size-dependent fashion. The new classification aims to standardize current diagnostic and management procedures and enable systematic and prognostically relevant patient stratification. PETs are graded into 1 of 3 tiers, either as well-differentiated neuroendocrine tumors or poorly-differentiated neuroendocrine carcinomas, on the basis of stage-pertinent features such as proven invasion or metastasis (5).
The grading system still remains controversial, but clear signs of malignancy include metastasis and local or extrapancreatic invasion. Other characteristics that appear helpful in determining prognosis are tumor size and functional status, necrosis, mitotic activity, perineural invasion and angioinvasion, and possibly CD44 isoform upregulated expression and cytokeratin 19 immunostaining (5,23). Peptide production detected in the serum or by immunohistochemistry is not a prognostic factor for nonfunctional PETs (3). Nuclear pleomorphism is also not a useful predictor; however some studies have demonstrated a correlation between overall nuclear grade and prognosis (24). The TNM system has proved to be highly predictive of patient outcome and is easy to combine with histologic and clinicopathologic parameters to classify pancreatic endocrine tumors into groups of increasing malignant potential (19,22).
Treatment
*O�UIF�QBTU �USFBUNFOU�PQUJPOT�GPS�1&5T�IBWF�CFFO�MJNJUFE �with hormonal treatment with octreotide (somatostatin analogues) as the primary therapeutic approach. Some PETs possess especially strong hormone receptors, such as somatostatin receptors and uptake hormones strongly (5). This avidity can assist in diagnosis and may make some tumors vulnerable to hormone targeted therapies. Although the optimal clinical management of PNETs involves a multidisciplinary approach, surgery remains the only curative treatment for early-stage disease. The
surgical treatment continues to evolve for PETs, but the best outcome occurs in those treated with total tumor resection. Surgery may also have a role in patients with advanced-stage disease, including those with hepatic metastases (25). Alternative therapeutic approaches applied to PETs, including chemotherapy, radiofrequency ablation, transarterial chemoembolization, biotherapy, polypeptide radionuclide receptor therapy, antiangiogenic therapy, and selective internal radiotherapy (7). Chemotherapeutic agents have been used with limited efficacy (less effective in well-differentiated tumors). Several agents have shown activity and combining several thearpies, particularly doxorubicin with streptozocin, is often more effective (26). Although marginally effective in well-differentiated PETs, cisplatin with etoposide is active in poorly-differentiated neuroendocrine cancers (5,26).
Targeted therapy has a clear role as these tumors do PWFSFYQSFTT�SFDFQUPST�GPS�&(' �1%(' �*('�� �BOE�7&('��3FDFOU�TUVEJFT�EFNPOTUSBUF�1*�,�"LU�N503�QBUIXBZ� JT�involved in the pathogenesis of PETs (8,9). Based on the QIBTF�***�DMJOJDBM�USJBMT�EBUB �N503�JOIJCJUPS�&WFSPMJNVT�s igni f icant ly improved progress ion-free surviva l among patients with progressive advanced pancreatic neuroendocrine tumors as compared with placebo (9). This targeted chemotherapy agents have been approved by FDA in patients with progressive unresectable, locally advanced or metastatic pancreatic neuroendocrine tumors. The DPNCJOBUJPO�PG�BO�N503�JOIJCJUPS�BOE�B�7&('�JOIJCJUPS�has also showed promising results (8).
Conclusions
*O� TVNNBSZ � QBODSFBUJD� OFVSPFOEPDSJOF� UVNPST� BSF�generally indolent neoplasms, even though the majority do present at an advanced stage. Once PETs is suspected based on the histologic features, immunohistochemistry plays a critical role to confirm the diagnosis. The 2010 WHO classification of tumors of the digestive system introduces grading and staging tools for pancreatic neuroendocrine neoplasms. A carcinoid is now defined as a grade 1 or 2 neuroendocrine tumor and grade 3, small-cell or large-DFMM�DBSDJOPNBT�BSF�EFòOFE�BT�OFVSPFOEPDSJOF�DBSDJOPNBT��
Classification/Grade Mitotic count (per 10 hpf) Ki-67 Index (%)
NET-G1 <2 <3
NET-G2 2-20 3-20
NEC-G3 >20 >20NET, neuroendocrine tumor; NEC, neuroendocrine carcinoma; hpf, high power field; 10 HPF =2 mm2, at least 40 fields (at 400× magnification) evaluated in areas of highest mitotic density
Well-differentiated = G1 and G2 Poorly-differentiated= G3
Kulke, et al. JCO, 2011; 29: 934-943 Moran et al. Am J Clin Pathol, 2009; 131: 206-221
5/1/15
7
SURVIVAL AFTER RESECTION OF PNET BY GRADE (N=926)
13 Rindi et al. JNCI 2014
EVOLVING CLASSIFICATION SCHEME
Kulke, et al. JCO, 2011; 29: 934-943 Moran et al. Am J Clin Pathol, 2009; 131: 206-221
• Low- and intermediate grade tumors (G1/2) are grouped as well-differentiated NET
• Reference to grade or differentiation required in path report (Ki 67 or mitotic rate)
• NOT black and white—spectrum in terms of behavior
• Doesn’t account for discordance between K67/mitotic rate/morphology • Scheme evolving • Core biopsy ideal
• FNA may be OK for pancreas (McCall, et al. 2013)
• Rebiopsy if behavior not consistent with path • Potential for heterogeneity
• Cut-off for true high grade unclear • Large cell/small cell, Ki 67>55%? • Well diff morphology, Ki 67 20-55%?
5/1/15
8
DECISION TREE: NET
NET
Well-differentiated
“carcinoid”
PNET
Paraganglioma/pheochromocytoma
Poorly-differentiated Platinum-based
CTx
UNKNOWN PRIMARY IN ≥15%: DOES IDENTIFYING TYPE OF TUMOR MATTER? YES
§ Prevents complications (SB tumors)
§ Treatment options and response to treatment depends on primary site o Everolimus/sunitinib for PNET o Chemotherapy for PNET?
§ Incidence has increased over last several decades o 0.2 to 0.4 per 100,000 per year o ≈1.5% pancreatic cancers (≈400-1200 cases/yr) o Increases with age; peaks in 7th decade, median age ≈60 o 10% of all pancreatic cancers by prevalence
§ Arise from islets of Langerhans (usually well-differentiated)
§ Evenly distributed between head, body and tail
§ Most are well-differentiated (<10% G3)
Halfdanarson et al. Annals of Oncology, 2008; 19: 1727-33: Yao et al. JCO, 2008; 26: 3063; *Yao et al. Ann Surg Oncol, 2007; 14: 3492-3500; Strosberg et al. Pancreas 2009; 38: 255-58
PANCREATIC NET (PNET)
§ Most ( ≈70%) are nonfunctional o Functional:insulinoma >gastrinoma >glucagonomas
>VIPomas> somatostatinomas>others § 90% of insulinomas are benign (excellent outcome) § Insulinomas and gastrinomas more common in younger pt § Glucagonomas and serotonin-producing tumors more
common in elderly
o Pancreatic polypeptide production common and asymptomatic (nonfunctional)
§ 55-70% metastatic at diagnosis o Median OS stage IV 2-6 yr
Halfdanarson et al. Annals of Oncology, 2008; 19: 1727-33: Yao et al. JCO, 2008; 26: 3063; *Yao et al. Ann Surg Oncol, 2007; 14: 3492-3500; Strosberg et al. Pancreas 2009; 38: 255-58
5/1/15
10
Multiple endocrine neoplasia type 1
The most frequent inherited syndrome associated withthe occurrence of PETs is multiple endocrine neoplasiatype 1 (MEN1), an autosomal dominant disorder(incidence 1:20 000–40 000) clinically defined by thepresence of two or more of gastroenteropancreaticneuroendocrine tumours, parathyroid gland adenomaand hyperplasia and pituitary adenomas, with otherneoplastic lesions occurring occasionally (Lemos &Thakker 2008). About 10% of PETs occur as a partof the MEN1 syndrome, which is the result of aninactivating mutation of a TSG located on chromosome11q13 (Metz & Jensen 2008). Ten exons constitutetheMEN1 gene, and the corresponding protein, namedMENIN, is involved in different functions, such asinactivation of transcription factors at a nuclear level(JUND and SMAD3), modulation of cell cycle inhibitors(enhancing p27KIPI and p18INK4c function) andinteraction with DNA repair machinery. The finalresult is a negative control of the cell cycle (Karniket al. 2005). The recent description of the crystal struc-ture of MENIN allowed an improved description of itsinteraction with JUND and MLL (Huang et al. 2012).
The mutations spectrum of MEN1 is wide: about1300 different germline mutations have been describedand 10–12% of them occurred even in the absenceof a family history. How mutations lead to cancer isnot well understood, and the exact role of MENIN as atumour suppressor is controversial (Jensen et al. 2008).
Gene mapping in MEN1 patients have shown loss ofheterozygosity (LOH) in 50% of cases and that MEN1carcinogenesis is sustained by a typical Knudson’stwo-hit mechanism. LOH of MEN1 and other somaticmutations on the wild-type allele work as a second hitafter a first germline mutation.
PET patients with MEN1 are clinically similar tosporadic cases except for their earlier age at diseasepresentation. Neoplasms are non-functioning (NF)-PETs in about 80% of cases, gastrinomas in 54%,insulinomas in 15–20%, glucagonoma in 3% and rarelyVIPomas or GRFomas. NF-PETs are found micro-scopically in 80–100% of cases, !15% being sympto-matic (Jensen et al. 2008). Although 20–60% of MEN1patients have Zollinger–Ellison syndrome, gastrinomasarise far more frequently in the duodenum than in thepancreas (Metz & Jensen 2008).
Apart from the 10% associated with the typical MEN1syndrome, a high percentage of sporadic PETs presentwith molecular abnormalities of the MEN1 gene or ofits function, thus suggesting its crucial role in theirpathogenesis. Mutation of MEN1 and allelic loss ofchromosome 11q are the most common geneticalterations, especially among NF-PETs (Moore et al.2001a, Perren et al. 2007). Mutations of MEN1 havebeen found in 30% of sporadic NF-PETs, 7% ofinsulinomas, 36% of gastrinomas, 67% of glucagon-omas and 44% of VIPomas (Moore et al. 2001a). Recentdata from Jiao et al. (2011) report 44% of inactivatingmutations in MEN1 in a series of sporadic PETs.
Table 1 Genetic syndromes associated with inherited pancreatic endocrine tumours, including clinical features and molecular defects
Syndrome Gene
Gene functionmain molecularconsequences Major clinicals features
Patientswith PET(%)
PETsubtype
MetastaticPET (%)
Multiple endocrineneoplasia type I
Menin(11q13)
Oncosuppressor Two or more between: 20–100% 100% NF !10%Deregulation ofJunD, SMAD3
a) GEP-NET 54% Gastrinomasb) Parathyroid adenomas 15% Insulinomas
p27KIPI p18Ink4c c) Pituitary adenoma 3% Glucagonomas1% GFRomas or
VIPomasvon Hippel–
Lindau diseaseVHL
(3p25–26)Oncosuppressor One or two between: 5–17% 80–100% NF !10%Overexpressionof HIF andVEGF
a) Retinal or cerebellarhemangioblastomas
b) Renal cell carcinomac) Pheochromocytoma
von Recklinghausen’sdisease
NF1(17q11.2)
Oncosuppressor a) Cafe-au-lait skin spots Rare Duodenal somatos-tinomas (1–10%)
–
Deregulation ofRas pathway(mTOR)
b) Neurofibromas of anytype and localisation
Insulinomas (!1%)
Tuberous sclerosiscomplex
TSC1(9q34)
Oncosuppressor a) Skin alterations Very rare Mainly NF –b) Renal angiomyolipomas
Journal of Molecular Endocrinology (2012) 49, R37–R50 www.endocrinology-journals.org
Capurso, et al. J Molecular Endocrinology, 2012
MEN1
§ Most common neoplasm is parathyroid hyperplasia (98% of patients), followed panNET(50+%), pituitary adenomas (35%), and/or lung/thymus carcinoid tumors (10%). o NF tumors most common PNET o Gastrinomas and insulinomas are the most
common functional tumors o Often multiple
§ Type 2 gastric carcinoid tumors common in MEN1
patients with gastrinoma.
5/1/15
11
MEN1
§ Adrenal tumors also increased in MEN1
§ HyperPTH usually treated before PNET in patients with MEN1
§ Role of surgery for multifocal PNET controversial in MEN1 o Symptomatic functional tumors refractory to medical management o Tumor larger than 1–2 cm in size o Tumor with relatively rapid rate of growth over 6–12 months o Endoscopy with EUS is recommended prior to pancreatic surgery
• DAXX/ATRX (43%), MEN1 (44%), and mTOR (14%) pathway genes are frequently altered in sporadic PNET
able tumors or extensive metastatic disease, andmedical therapies are relatively ineffective inthese cases.
There is currently insufficient informationabout this tumor to either predict prognosis ofpatients diagnosed with PanNETs or to devel-op companion diagnostics and personalizedtreatments to improve disease management. Bial-lelic inactivation of the MEN1 gene, usuallythrough a mutation in one allele coupled with
loss of the remaining wild-type allele, occursin 25 to 30% of PanNETs (5, 6). MEN1 is a tu-mor suppressor gene that, when mutated inthe germline, predisposes to multiple endocrineneoplasia type 1 syndrome. Chromosomal gainsand losses and expression analyses have re-vealed candidate loci for genes involved in thedevelopment of PanNETs, but these have notbeen substantiated through genetic or functionalanalyses (7–9).
To gain insights into the genetic basis of thistumor type, we determined the exomic sequenceof ~18,000 protein-coding genes in a discoveryset of 10 well-characterized sporadic PanNETs.A clinically homogeneous set of tumors of highneoplastic cellularity is essential for the success-ful identification of genes and pathways involvedin any tumor type. Thus, we excluded small-cell and large-cell neuroendocrine carcinomasand studied only samples that were not part ofa familial syndrome associated with PanNETs(table S1) (1). We microdissected tumor sam-ples in order to achieve a neoplastic cellularityof >80%. DNA from the enriched neoplastic
samples and from matched non-neoplastic tissuefrom 10 patients was used to prepare fragmentlibraries suitable for massively parallel sequenc-ing. The coding sequences were enriched bycapture with the SureSelect Enrichment Systemand sequenced by use of an Illumina GAIIx plat-form (10). The average coverage of each base inthe targeted regions was 101-fold, and 94.8% ofthe bases were represented by at least 10 reads(table S2).
We identified 157 somatic mutations in 149genes among the 10 tumors used in the discoveryset. Themutations per tumor ranged from 8 to 23,with a mean of 16 (table S3). Of these mutations,91% were validated by means of Sanger sequenc-ing. There were some obvious differences be-tween the genetic landscapes of PanNETs andthose of pancreatic ductal adenocarcinomas(PDAC) (11). First, there were 60% fewer genesmutated per tumor in PanNETs than in PDACs.Second, the genes most commonly affected bymutation in PDACs (KRAS, TGF-b pathway,CDKN2A, and TP53) were rarely altered inPanNETs and vice versa (Table 1). Third, the
Table 1. Comparison of commonly mutated genesin PanNETs and PDAC based on 68 PanNETs and114 PDACs.
4 MARCH 2011 VOL 331 SCIENCE www.sciencemag.org1200
REPORTS
on
May
10,
201
3w
ww
.sci
ence
mag
.org
Dow
nloa
ded
from
spectrum of mutations in PDAC and PanNETwere different, with C-to-T transitions more com-mon in PDACs than in PanNETs and C-to-Gtransversions more common in PanNETs than inPDACs (table S4). This suggests that mutationsin PanNETs and PDAC arise through differentmechanisms, perhaps because of exposure todifferent environmental carcinogens or throughthe action of different DNA-repair pathways.
We next selected genes for further analysisthat were well-documented components of apathway that was genetically altered in morethan one tumor because alterations in these genesare most likely to be clinically relevant. Fourgenes were mutated in at least two tumors inthe discovery set:MEN1 in five, DAXX in three,PTEN in two, and TSC2 in two. ATRX was mu-tated in only one sample in the discovery set,but its product forms a heterodimer with DAXXand, therefore, is part of the same pathway, so itwas also evaluated in the validation set. Sim-ilarly, PIK3CAwas included because its productis part of the mammalian target of rapamycin(mTOR) pathway that includes PTEN and TSC2(12–14). The sequences of these genes werethen determined by means of Sanger sequenc-ing in a validation set consisting of 58 additionalPanNETs and their corresponding normal tis-sues (Fig. 1, A and B). In total, somatic muta-
tions in MEN1, DAXX, ATRX, PTEN, TSC2, andPIK3CAwere identified in 44.1%, 25%, 17.6%,7.3%, 8.8%, and 1.4% PanNETs, respectively(Table 2).
Of the 30 mutations in MEN1, 25 were in-activating mutations [18 insertions or deletions(“indels”), 5 nonsense, and 2 splice-site muta-tions], whereas five were missense. At least 11were homozygous; in the others, the presence of“contaminating”DNA from normal cells made itdifficult to reliably distinguish heterozygous fromhomozygous changes. MEN1 encodes menin, anuclear protein that acts as a scaffold to regulategene transcription by coordinating chromatin re-modeling. It is an essential component of theMLLSET1–like histonemethyltransferase (HMT)complex (15–19). Overall,MEN1wasmutated in30 of the 68 PanNETs used in the discovery andvalidation sets combined.
DAXX and ATRX were mutated in 17 and 12PanNETs, respectively. No tumor with a mutationin DAXX had a mutation in ATRX, which is con-sistent with their presumptive function withinthe same pathway. Overall, 29 of 68 PanNETs(42.6%) had a mutation in this pathway. Therewere 11 indels and four nonsense mutations inDAXX and six indels and three nonsense mu-tations in ATRX. The three ATRX missensemutations were within the conserved helicase
domain, whereas the DAXX missense mutationswere nonconserved changes. Five DAXX andfour ATRX mutations were homozygous, indi-cating loss of the other allele. The high ratioof inactivating-to-missense mutations in bothgenes establishes them as PanNET tumor sup-pressor genes. Loss of immunolabeling forDAXX and ATRX correlated with mutation ofthe respective gene (fig. S1, A and B, and tableS5). From these data, we hypothesize that bothcopies of DAXX are generally inactivated, onethrough mutation and the other either throughloss of the nonmutated allele or epigenetic si-lencing. We also hypothesize that both copiesof ATRX are inactivated, one through mutationand the other through chromosome X inactiva-tion. Recently, it has been shown that DAXX isan H3.3-specific histone chaperone (20). ATRXencodes for a protein that at the amino-terminushas an ADD (ATRX-DNMTT3-DNMT3L) do-main and a carboxy-terminal helicase domain.Almost all missense disease-causing mutationsare within these two domains (21). DAXX andATRX interact, and both are required for H3.3incorporation at the telomeres; ATRX is alsorequired for suppression of telomeric repeat–containing RNA expression (22–24). ATRX wasrecently shown to target CpG islands and G-richtandem repeats (25), which exist close to telomericregions.
We identified five PTEN mutations, twoindels and three missense; six TSC2 mutations,one indel, one nonsense, and four missense; andone PIK3CAmissense mutation. Previously pub-lished expression analyses have indicated that theexpression of genes in the mTOR pathway isaltered in most PanNETs (26, 27). Our datasuggest that, at least at the genetic level, only asubset of PanNETs have alterations of thispathway. This finding may have direct clinicalapplication through prioritization of patients fortherapy with mTOR pathway inhibitors. Ever-olimus [also called Afinitor, RAD-001, and 40-O-(hydroxyethyl)-rapamycin] has been shown toincrease progression-free survival in a subset ofPanNET patients with advanced disease (28). Ifthe mutational status of genes coding for proteinsin the mTOR pathway predicts clinical responseto mTOR inhibitors, it should be possible toselect patients who would benefit most from anmTOR inhibitor through analysis of these genes inpatients’ tumors (29, 30).
All 68 tumors evaluated in this study werefrom patients undergoing aggressive intervention(table S6) and included patients undergoingcurative resection as well as those with metastaticdisease. Mutations in MEN1, DAXX/ATRX, orthe combination of bothMEN1 andDAXX/ATRXwere associated with prolonged survival relativeto those patients whose tumors lacked these muta-tions (Fig. 1, C and D, and table S7). This wasparticularly evident in patients with metastaticdisease and with mutations in both MEN1 andDAXX/ATRX: 100% of patients with PanNETsthat had these mutations survived at least 10 years,
Fig. 1. (A and B) Examples of traces showing mutations in DNA isolated from cancer cells [(A) and (B),bottom] but not from normal cells of the same patient [(A) and (B), top]. (C and D) Kaplan-Meier plots ofoverall survival of patients with metastatic PanNETs. (C) Fifteen patients with a DAXX or ATRX genemutation versus 12 patients in whom both genes were wild type (WT) [hazard ratio 0.22, 95% confidenceinterval (CI) 0.06 to 0.84, P = 0.03]. (D) Nine patients with mutations in MEN1 as well as either DAXX orATRX versus six patients in which all three genes were WT (hazard ratio 0.07, 95% CI 0.009 to 0.53, P =0.01). All types of mutations in these genes were included in the analysis.
4 MARCH 2011 VOL 331 SCIENCE www.sciencemag.org1202
REPORTS
on
May
10,
201
3w
ww
.sci
ence
mag
.org
Dow
nloa
ded
from
Jiao, et al. Science, 2011
5/1/15
12
PNET GENETICS
§ Prognostic value of specific mutations unclear o Conflicting data
§ Predictive value of specific alterations unclear
23
RESECTABLE DISEASE Pancreatic NETs
5/1/15
13
WORK-UP
§ Multiphasic CT or MRI § CGA (elevated in 60% of PNET) § As appropriate: somatostatin scintigraphy, EUS, +/- additional
biochemical evaluation o Gastrin o Insulin, proinsulin ,c-peptide, FBS o Glucagon o Pancreatic polypeptide o VIP o Somatostatin o Calcitonin o PTHrP o RARE: GHRH, ACTH, 24 hr urine 5HIAA, LH, renin, IGF2, etc
Kulke, et al. Pancreas, 2010
TREATMENT: SURGICAL RESECTION
• Type of surgery depends on location of tumor • Enucleation • Distal pancreatectomy • Central pancreatectomy • Whipple (pancreatic duodenectomy) • Total pancreatecomy
• Curative surgery, TNM stage and grade are predictors of survival
• No role for adjuvant therapy
5/1/15
14
PNET SURVIVAL BY STAGE (N=926)
Rindi et al. JNCI 2014
METASTATIC DISEASE
5/1/15
15
LIVER-‐DOMINANT METASTATIC DISEASE (NO RCT) § Resection of known disease if possible (≥90%)
o Median OS 10 yr (retrospective 339 pt)
§ Benefits greatest if <25% liver involvement or symptomatic high volume disease (vs hepatic arterial therapy)
§ Recurrence 95% at 5 yr, 99% at 10 yr n Mayo, et al. Ann Surg Onc, 2010;17
o 10 year OS 50% if RO or R1 resection (172 pt) • 15% R1; 10% RFA alone; 13% RFA plus surgery • Thoracic primary associated with worse outcome on
multivariate analysis (less so for late recurrence) • Late recurrence an issue (annual scan for life?)
n Glazer, et al. HBP (Oxford), 2010
ADVANCED UNRESECTABLE DISEASE
Pancreatic NETs
5/1/15
16
31
Well-differentiated NET
of hormone-mediated symptoms
Progressive disease (need for anti-tumor effect)
UNRESECTABLE DISEASE: INDICATIONS FOR THERAPY
Patient selection is key!
Ability to resect all known disease
SOMATOSTATIN (SST) § Bioactive neuropeptide
o PreproèSST-14 and SST-28 (Short t ½)
§ Produced and acts locally: o êGlandular and exocrine secretions:
§ Inhibits GH/ACTH/TSH release § Pan-inhibitor of GI tract hormone release (insulin, glucagon) § Inhibits release of gastric acid, amylase
o Antiproliferative
§ Mediates inhibitory effects thru 5 GCPR (SSTR 1-5) o Expressed throughout CNS, GI tract, endocrine/exocrine glands, &
immune/inflammatory cells
Schmid et al. Mol Cell Endocrinol 2008;286:69–74;
5/1/15
17
REGULATORY ACTION OF SOMATOSTATIN RECEPTORS
Action SST1 SST2 SST3 SST4 SST5
Antisecretory X X X Anti-angiogenic X X X Antiproliferative/ Inhibit cell cycle (G1 arrest) X X X X
Induction of apoptosis X X X
Adapted from Susini C, Buscail L and Weckbecker G, Lewis I, Albert R, et al.1 References: 1. Weckbecker G, Lewis I, Albert R, et al. Nature Rev Drug Discov. 2003, 2:999-1017. 2. Öberg K, Kvols
L, Caplin M, et al. Ann Oncol. 2004; 15:966-973. 3. Susini C, Buscail L. Ann Oncol. 2006; 17:1733-1742.
Effects probably SSTR subtype- and tissue-specific; may reflect interaction between two or more receptor types (e.g. SSTR subtypes and/
or other GPCR, eg D2R)
SOMATOSTATIN ANALOGS (SSTA) AND NETS
§ SSTa indicated for the treatment of hormone-‐mediated sx1:
§ Octreo@de (SQ)/Octreo@de LAR (IM q mo) o Sandosta@n®, Sandosta@n LAR® o Approved for acromegaly and carcinoid syndrome
§ Lanreo@de (SQ q 14d)/Lanreo@de autogel ( SQ q mo) o Approved for acromegaly (Somatuline) and for tumor control of well diff NETS in US (including PNETs)
Option for open-label lanreotide on extension study
Caplin, et al. NEJM, 2014
5/1/15
26
CLARINET: PFS BY CENTRAL REVIEW (1°EP)
T h e n e w e ngl a nd j o u r na l o f m e dic i n e
n engl j med 371;3 nejm.org july 17, 2014228
according to RECIST in the 3 to 6 months before randomization (96%).
EfficacyProgression-free Survival (Primary End Point)More patients in the placebo group than in the lanreotide group had centrally assessed disease-progression events (58 vs. 30 patients), and 2 pa-tients in each group died. Progression-free sur-vival was significantly prolonged with lanreotide as compared with placebo in the primary analy-sis (median progression-free survival, not reached vs. 18.0 months, P<0.001 by the stratified log-rank test; hazard ratio for progression or death with lanreotide vs. placebo, 0.47; 95% confidence interval [CI], 0.30 to 0.73) (Fig. 1). At 24 months, the estimated rates of progression-free survival were 65.1% (95% CI, 54.0 to 74.1) in the lanreo-tide group and 33.0% (95% CI, 23.0 to 43.3) in the placebo group. All supportive and sensitiv-ity analyses corroborated the primary analysis (Table S1 in the Supplementary Appendix).
Hazard ratios for disease progression or death generally favored lanreotide over placebo in the predefined subgroups (Fig. 2, and Fig. S2
Table 1. Baseline Demographic and Disease Characteristics of the Patients (Intention-to-Treat Population).*
* Plus–minus values are means ±SD. Additional baseline data are provided in Table S3 in the Supplementary Appendix. Post hoc analyses confirmed that there were no significant between-group differences at baseline. The midgut was de-fined as the small intestine and appendix, and the hindgut was defined as the large intestine, rectum, anal canal, and anus.
† Two patients in each group had gastrinomas.‡ Ki-67 thresholds for the tumor grade index were based on the World Health Organization 2010 classification.16 Patients
who had Ki-67 values greater than 2% and up to 10% in the present study were classified as having grade 2 disease.
Patie
nts
with
Pro
gres
sion
-free
Surv
ival
(%)
100
80
90
70
60
40
30
10
50
20
00 63 9 12 18 24 27
Months
P<0.001 for the comparison of progression-free survivalHazard ratio for progression or death, 0.47 (95% CI, 0.30–0.73)
No. at RiskLanreotidePlacebo
101103
94101
8487
7876
7159
6143
4026
00
Lanreotide 120 mg32 events, 101 patientsMedian not reached
Placebo60 events, 103 patientsMedian, 18.0 mo (95% CI, 12.1–24.0)
Shown are estimates of progression-free survival among patients who re-ceived lanreotide at a dose of 120 mg and patients who received placebo. Kaplan–Meier curves were compared with the use of a stratified log-rank test, with stratification according to the presence or absence of tumor pro-gression at baseline and the receipt or nonreceipt of previous therapy. The hazard ratio was derived from a Cox proportional-hazards model with terms for study treatment, the presence or absence of tumor progression at base-line, and the receipt or nonreceipt of previous therapy.
The New England Journal of Medicine Downloaded from nejm.org at UC SHARED JOURNAL COLLECTION on September 4, 2014. For personal use only. No other uses without permission.
T h e n e w e ngl a nd j o u r na l o f m e dic i n e
n engl j med 371;3 nejm.org july 17, 2014228
according to RECIST in the 3 to 6 months before randomization (96%).
EfficacyProgression-free Survival (Primary End Point)More patients in the placebo group than in the lanreotide group had centrally assessed disease-progression events (58 vs. 30 patients), and 2 pa-tients in each group died. Progression-free sur-vival was significantly prolonged with lanreotide as compared with placebo in the primary analy-sis (median progression-free survival, not reached vs. 18.0 months, P<0.001 by the stratified log-rank test; hazard ratio for progression or death with lanreotide vs. placebo, 0.47; 95% confidence interval [CI], 0.30 to 0.73) (Fig. 1). At 24 months, the estimated rates of progression-free survival were 65.1% (95% CI, 54.0 to 74.1) in the lanreo-tide group and 33.0% (95% CI, 23.0 to 43.3) in the placebo group. All supportive and sensitiv-ity analyses corroborated the primary analysis (Table S1 in the Supplementary Appendix).
Hazard ratios for disease progression or death generally favored lanreotide over placebo in the predefined subgroups (Fig. 2, and Fig. S2
Table 1. Baseline Demographic and Disease Characteristics of the Patients (Intention-to-Treat Population).*
* Plus–minus values are means ±SD. Additional baseline data are provided in Table S3 in the Supplementary Appendix. Post hoc analyses confirmed that there were no significant between-group differences at baseline. The midgut was de-fined as the small intestine and appendix, and the hindgut was defined as the large intestine, rectum, anal canal, and anus.
† Two patients in each group had gastrinomas.‡ Ki-67 thresholds for the tumor grade index were based on the World Health Organization 2010 classification.16 Patients
who had Ki-67 values greater than 2% and up to 10% in the present study were classified as having grade 2 disease.
Patie
nts
with
Pro
gres
sion
-free
Surv
ival
(%)
100
80
90
70
60
40
30
10
50
20
00 63 9 12 18 24 27
Months
P<0.001 for the comparison of progression-free survivalHazard ratio for progression or death, 0.47 (95% CI, 0.30–0.73)
No. at RiskLanreotidePlacebo
101103
94101
8487
7876
7159
6143
4026
00
Lanreotide 120 mg32 events, 101 patientsMedian not reached
Placebo60 events, 103 patientsMedian, 18.0 mo (95% CI, 12.1–24.0)
Shown are estimates of progression-free survival among patients who re-ceived lanreotide at a dose of 120 mg and patients who received placebo. Kaplan–Meier curves were compared with the use of a stratified log-rank test, with stratification according to the presence or absence of tumor pro-gression at baseline and the receipt or nonreceipt of previous therapy. The hazard ratio was derived from a Cox proportional-hazards model with terms for study treatment, the presence or absence of tumor progression at base-line, and the receipt or nonreceipt of previous therapy.
The New England Journal of Medicine Downloaded from nejm.org at UC SHARED JOURNAL COLLECTION on September 4, 2014. For personal use only. No other uses without permission.
Lanreotide in Enteropancreatic Neuroendocrine Tumors
n engl j med 371;3 nejm.org july 17, 2014 231
Randomized Study on the Effect of Octreotide LAR in the Control of Tumor Growth in Patients with Metastatic Neuroendocrine Midgut Tumors [PROMID]).11 Although PROMID showed a sig-nificantly prolonged time to tumor progression with octreotide long-acting release therapy as compared with placebo in patients with midgut tumors, the study was narrowly focused so that it was made up almost entirely of patients with grade 1 tumors and few patients had hepatic tumor volumes greater than 10%.11 Our study showed an antiproliferative effect of lanreotide in a study population of patients with entero-pancreatic tumors, with Ki-67 values extending into grade 2 (Ki-67 <10%) and with larger he-patic tumor volumes. Although detailed compari-sons of the findings from these two studies must be made with caution, it seems likely that the patients in our study, as compared with those in PROMID, had, in general, more indolent tumors, since the median time to diagnosis was longer in our study than in PROMID (14.7 months vs. 4.3 months). In addition, almost all the patients in our study had stable disease at baseline, and it seems unlikely, considering the shorter time to disease progression in the PROMID placebo group than in the CLARINET placebo group, that PROMID had such a predominance of pa-tients with stable disease. How ever, since disease-progression status was not documented in PROMID, this cannot be confirmed. Moreover, disease progression was assessed differently in the two studies: the bidimensional WHO criteria were used in PROMID and the unidimensional RECIST, version 1.0, criteria were used in our study. Since a bidimensional measure shows a larger percentage increase in tumor size than a uni-dimensional measure of the same tumor re-sponse, WHO-based assessments might show a shorter progression-free survival. These differ-ences notwithstanding, the two studies are aligned in affirming clinically relevant antiproliferative effects with long-acting somatostatin analogues in patients with neuroendocrine tumors.
Current clinical practice guidelines regard-ing the use of somatostatin analogues for con-trol of advanced enteropancreatic neuroendo-crine tumors are based largely on findings from PROMID.5,21 However, in cases in which evi-dence from clinical trials is lacking or individual circumstances dictate, current guidelines sug-gest that a period of deferred treatment (a “wait
and see” policy) may be appropriate. The placebo group in our study may be considered a surro-gate for deferred treatment, and the long period of progression-free survival in this group may appear to support this approach. However, our study principally examined the prevention of dis-
Table 3. Adverse Events (Safety Population).*
EventLanreotide (N = 101)
Placebo(N = 103)
no. of patients (%)
Any adverse event 89 (88) 93 (90)
Any adverse event related to study treatment 50 (50) 29 (28)
Any adverse event according to intensity†
Severe 26 (26) 32 (31)
Moderate 44 (44) 44 (43)
Mild 17 (17) 17 (17)
Any serious adverse event 25 (25) 32 (31)
Serious adverse event related to study treatment‡
3 (3) 1 (1)
Withdrawal from study because of any adverse event§
3 (3) 3 (3)
Withdrawal because of adverse event related to study treatment
1 (1) 0
Study treatment–related adverse events in ≥5% of patients
Diarrhea 26 (26) 9 (9)
Abdominal pain 14 (14) 2 (2)
Cholelithiasis 10 (10) 3 (3)
Flatulence 8 (8) 5 (5)
Injection-site pain 7 (7) 3 (3)
Nausea 7 (7) 2 (2)
Vomiting 7 (7) 0
Headache 5 (5) 2 (2)
Lethargy 5 (5) 1 (1)
Hyperglycemia 5 (5) 0
Decreased level of pancreatic enzymes 5 (5) 0
* Adverse events were defined according to the Medical Dictionary for Regulatory Activities, version 16.0.
† For patients with multiple adverse events, events with the maximum intensity are shown; data are missing for two patients in the lanreotide group.
‡ There were seven events (hyperglycemia, diabetes mellitus, nausea, vomiting, abdominal pain, biliary fistula, and cholelithiasis) in the lanreotide group and one event (bile duct stenosis) in the placebo group.
§ Intestinal obstruction, sepsis, hypoglycemia, esophageal carcinoma, and cir-culatory collapse were not considered to be related to the study treatment. “Liver decompensation” (the term used by the investigator) was considered by the investigator to be related to the study treatment because of the timing of the event (the day after the first injection); the event was concurrent with an episode of food poisoning, and the patient recovered without sequelae after 3.5 months.
The New England Journal of Medicine Downloaded from nejm.org at UC SHARED JOURNAL COLLECTION on September 4, 2014. For personal use only. No other uses without permission.
§ Lanreotide improves PFS in nonfunctional well-diff NET (Ki 67<10%) Median PFS 18 mo v NR (P,0.001)
§ No Δ 2 yr OS o Extension study confounds interpretation
§ Excellent median PFS in control group argues for watch and wait
§ Enriched for patients with SD o 95% of patients had SD by RECIST in the 3-6 mo before study drug.
§ Value Ki 67>10%? Pt with PD?
§ FDA APPROVED 12/2014 to improve PFS of well- or moderately diff advanced GEP-NETs
54
5/1/15
28
TREATMENT OPTIONS: SYSTEMIC THERAPY
Peptide receptor radiotherapy (PRRT)
o Potential value (e.g. [177 Lu-DOTA 0,Tyr3]octreotate or 90Y-‐DOTATOC) in pt with high tumor uptake on somatostatin receptor scintigraphy § Other radioconjugates under study
o Symptom control, SD, and/or radiographic responses (4-30%) have been reported
o No prospective RCT evaluating toxicity/anti-tumor efficacy § DJ Kwekkeboom et al, J Clin Onc 2008; D Bushnell et al, J Clin Onc 2010
Boudreaux, et al. Pancreas, 2010; 39; 753-766.; Strosberg, et al. Cancer Control, 2011; 18: 127-137
• Well-differentiated NET: No accepted standard treatment
• Carcinoids: RR<20%
• PNET:
• inconsistent RR (6-40%) with streptozotocin-based therapy (Moertel, et al. 1992, NEJM; Cheng and Saltz, 1999, Cancer) • 70% RR in 1st line PNET (retrospective, n=30) with capecitabine/temozolomide (Strosberg, et al. Cancer 2011)
• Treatment related toxicity often limiting
5/1/15
29
TEMOZOLOMIDE-BASED CHEMOTHERAPY: SUMMARY
§ Activity in NET o May be dose/schedule/partner-dependent
§ 3+ regimens under study o May be MGMT-dependent
§ Definition of high vs. low? o May be disease-dependent: PNET vs carcinoid
§ 70% RR in 30 PNET (retrospective) with capecitabine/temozolomide • Strosberg, et al. Cancer 2011
o May be site-dependent (e.g. thymic vs bronchial vs SB)
§ Toxicity may be schedule-dependent § Utility of MGMT testing needs prospective validation
§ Prospective, randomized trials are needed! § Not approved for this indication
ECOG 2211: RANDOMIZED PHASE II STUDY OF CAPECITABINE/TEMOZOLOMIDE VS TEMOZOLOMIDE ALONE IN ADVANCED, PROGRESSIVE,
WELL-DIFFERENTIATED PNET
RANDOM I ZAT I ON N=145
Eligibility criteria: • Well‑differenLated PNET • Disease progression in past 12 months • ECOG 0/1
§ NETs are highly vascular and express VEGF and its receptors
§ VEGF expression correlates with metastases and decreased progression free survival (PFS)
§ In preclinical models, VEGF is a valid target for therapy o VEGF RTK inhibitors active in RIPTAg model
Terris B, Scoazec JY, Rubbia L, et al. Histopathology 21: 133-8, 1998 Inoue M, Hager JH, Ferrara N. et al. Cancer Cell 1: 193-202, 2002 Zhang J, et al. Cancer, 2007:109; 1478-86. Konno H, et al. Jpn J Cancer Res, 1998:89; 933-9
• The treatment of advanced NET’s is controversial • Pla@num-‐based regimens are standard in the 1st line tx of
PDNET; no effec@ve salvage therapy exists 3
• Carcinoid tumors are oeen resistant to standard chemotherapy • The role of streptozocin-‐based chemotherapy in PNET is
controversial 4,5 • XELOX has ac@vity in NET’s based on phase II data6 • NET’s are highly vascular and express VEGF 1,2 • Preliminary data suggest that VEGF inhibitors have ac@vity in
NET’s 7,8
• Bevacizumab enhances the ac@vity of chemotherapy in metasta@c colorectal cancer 9,10
• The treatment of advanced NET’s is controversial • Pla@num-‐based regimens are standard in the 1st line tx of
PDNET; no effec@ve salvage therapy exists 3
• Carcinoid tumors are oeen resistant to standard chemotherapy • The role of streptozocin-‐based chemotherapy in PNET is
controversial 4,5 • XELOX has ac@vity in NET’s based on phase II data6 • NET’s are highly vascular and express VEGF 1,2 • Preliminary data suggest that VEGF inhibitors have ac@vity in
NET’s 7,8
• Bevacizumab enhances the ac@vity of chemotherapy in metasta@c colorectal cancer 9,10
• The treatment of advanced NET’s is controversial • Pla@num-‐based regimens are standard in the 1st line tx of
PDNET; no effec@ve salvage therapy exists 3
• Carcinoid tumors are oeen resistant to standard chemotherapy • The role of streptozocin-‐based chemotherapy in PNET is
controversial 4,5 • XELOX has ac@vity in NET’s based on phase II data6 • NET’s are highly vascular and express VEGF 1,2 • Preliminary data suggest that VEGF inhibitors have ac@vity in
NET’s 7,8
• Bevacizumab enhances the ac@vity of chemotherapy in metasta@c colorectal cancer 9,10
• The treatment of advanced NET’s is controversial • Pla@num-‐based regimens are standard in the 1st line tx of
PDNET; no effec@ve salvage therapy exists 3
• Carcinoid tumors are oeen resistant to standard chemotherapy • The role of streptozocin-‐based chemotherapy in PNET is
controversial 4,5 • XELOX has ac@vity in NET’s based on phase II data6 • NET’s are highly vascular and express VEGF 1,2 • Preliminary data suggest that VEGF inhibitors have ac@vity in
NET’s 7,8
• Bevacizumab enhances the ac@vity of chemotherapy in metasta@c colorectal cancer 9,10
• The treatment of advanced NET’s is controversial • Pla@num-‐based regimens are standard in the 1st line tx of
PDNET; no effec@ve salvage therapy exists 3
• Carcinoid tumors are oeen resistant to standard chemotherapy • The role of streptozocin-‐based chemotherapy in PNET is
controversial 4,5 • XELOX has ac@vity in NET’s based on phase II data6 • NET’s are highly vascular and express VEGF 1,2 • Preliminary data suggest that VEGF inhibitors have ac@vity in
NET’s 7,8
• Bevacizumab enhances the ac@vity of chemotherapy in metasta@c colorectal cancer 9,10
5/1/15
31
PHASE 3, RANDOMIZED, DOUBLE-BLIND STUDY OF SUNITINIB VS PLACEBO IN PATIENTS WITH ADVANCED, PROGRESSIVE, WELL-
DIFFERENTIATED PNET
RANDOM I ZAT I ON N=340 planned
(171 actual) Closed early
Eligibility criteria: • Well‑differenLated PNET • Disease progression in past 12 months • ECOG 0/1
SuniLnib 37.5 mg PO Q day w/o breaks
Placebo*
1:1 1°EP: PFS
*With best supporLve care
SomatostaLn analogs permiled in both arms
After trial closure all patients became candidates for open-label sunitinib in trial NCT00443534 or NCT00428220
Raymond, et al. NEJM, 2011
PROGRESSION-FREE SURVIVAL (PRIMARY ENDPOINT)
62
1.0
0.8
0.6
0.4
0.2
0
Prop
orLo
n of paL
ents
0 5 10 15 20 25
86 39 19 4 0 85 28 7 2 1 0
Number at risk SuniLnib Placebo
Time, months
Median PFS SuniLnib 11.4 months (95% CI 7.4, 19.8) Placebo 5.5 months (95% CI 3.6, 7.4)
HR=0.418 (95% CI 0.263, 0.662) P<.001
Raymond E, et al. N Engl J Med. 2011 HR, hazard ratio.
5/1/15
32
SUMMARY
§ Sunitinib improves PFS in progressive, well-differentiated pNET o Benefit across subgroups (including on-study and prior SSA) o PR rate low (9%) o QOL preserved
§ Hypertension, fatigue, diarrhea, nausea o Overall survival data should be interpreted with caution (70% X-over)
§ Longer f/u needed o FDA-approved for progressive PNET indication in May 11’
§ Unanswered questions:
o Optimal role/sequence in context of other “available” therapies (e.g. SIRT, everolimus, PRRT, chemotherapy) ?
o Value in other low-grade NET (e.g. carcinoid)?
MTOR PATHWAY INHIBITOR Advanced PanNETs
5/1/15
33
MTOR PATHWAY AS A TARGET FOR THERAPY IN NET
1. Bjornsti M-A, Houghton PJ. Nature Rev Cancer. 2004;4:335-348. 2. von Wichert G, Jehle PM, Hoeflich A, et al. Cancer Res. 2000;60:4573-4581.
• mTOR has a central role in a number of proliferation pathways1
• Abnormalities in the mTOR pathway are involved in the development of some NETs2
• -TSC-2 is negative regulator of mTOR
• mTOR components mutated in 15%
• mTOR signaling promotes cell metabolism, angiogenesis, and cell proliferation1,2
• mediator of VEGF and IGF-1 signaling
RADIANT-3: EVEROLIMUS VS. PLACEBO IN ADVANCED PANCREATIC NET
Everolimus 10 mg/d + best supportive care*
Placebo + best supportive care*
Treatment continued
until progression
Random
ize
Patients with advanced pNET
PD w/i 12 mo
n=410*
Cross over 1:1
*concurrent somatostatin analogs allowed Randomization Aug. 2007 – May. 2009
1° EP PFS (inv-reported)
Yao, et al. NEJM, 2011
5/1/15
34
PFS BY INVESTIGATOR REVIEW
• P-value obtained from stratified one-sided log rank test • Hazard ratio is obtained from stratified unadjusted Cox model
Hazard ratio = 0.35; 95% CI [0.27-0.45] P-value: <0.0001
0 1
0 0
Time (months)
100
80
Per
cent
age
even
t-fre
e
Censoring Times Everolimus (n/N = 109/207) Placebo (n/N = 165/203)
60
40
20
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Yao, et al. NEJM, 2011
SUMMARY: EVEROLIMUS IN PNET
• Everolimus therapy resulted in a significant 6.4 month increase in median PFS
– 4.6 months v 11.0 months
– No OS benefit
• Stability more common than significant shrinkage
§ 4.8% RR with RAD vs 2% with placebo
§ Everolimus has an acceptable safety profile
o Mouth sores, rash, hyperglycemia, hyperlipidemia, hypophosphatemia
§ Approved by FDA for progressive PNET in May 11’
§ Relationship between response and mutational status unknown
5/1/15
35
EFFICACY IN SUNITINIB OR EVEROLIMUS IN PNET RANDOMIZED TRIALS
Sunitinib (n=171)
Everolimus (n=410)
Median PFS 11.4 mos (vs. 5.5 mos in placebo
arm)
11.0 mos (vs 4.6 mos in placebo
arm)
Overall Response Rate (RECIST)
9.3% 5%
Partial Response or Stable Disease
72% 78%
Survival Advantage Demonstrated?
No* No*
*Pts receiving placebo in either study had opportunity to receive study drug following progression
Yao, et al. NEJM, 2011 Raymond, et al. NEJM, 2011
COMBINATION MTOR INHIBITOR +VEGF INHIBITOR
PH II TEMSIROLIMUS + BEVACIZUMAB IN PROGRESSIVE PANNETS
Confirmed PR 23 (41%) 6-month PFS 79% Med PFS 13.2 mo 12-month PFS 48%
Hobday. JCO 2014
Advanced, progressive G1/G2 pancrea@c NETs
(n=55)
Temsirolimus 25mg IV days 1, 8, 15, 22 Bevacizumab 10 mg/kg days 1, 15
Repeat Q28 days Primary EP: RR and 6-‐month PFS
Most common grade 3 to 4 adverse events: hypertension (21%), fatigue (16%), lymphopenia (14%), and hyperglycemia (14%).
5/1/15
36
CALGB 80701: RANDOMIZED PHASE II STUDY OF EVEROLIMUS ALONE OR IN COMBINATION WITH BEVACIZUMAB, IN PATIENTS WITH ADVANCED
PANCREATIC NET (OPEN 10/2010, KULKE, PI)
138 pts
RANDOMI ZE
Arm 1: Everolimus 10 mg po qd + octreotide LAR
Arm 2: Everolimus 10 mg po qd + Bevacizumab 10 mg/kg IV q 2 wks + octreotide LAR
Primary Endpoint:
INV –report PFS
Opened: October 2010
Closed to accrual– results pending
SUMMARY: ADVANCED WELL-DIFF PANCREATIC NET
§ Liver resection associated with prolonged survival but is not curative
§ Asymptomatic patients with stable unresectable disease and low tumor burden can be observed and monitored
§ SSTa have documented antitumor activity in well diff NET
§ Sequential therapy with targeted agents (everolimus or sunitinib) in patients with symptoms, clinically significant tumor burden, and/or progressive PNET
§ Stabilization >>> shrinkage § Neither agent has been shown to improve overall survival § Optimal sequence unknown § No established role in carcinoid
§ Chemotherapy often considered when tumor response required or when patients have failed targeted agents (PNET>>CARC)
5/1/15
37
§ Major advances in understanding and treatment of NET in past 10 years § Treatment needs to be individualized, weighing the risks of therapy vs
poten@al benefit o Predic@ve markers are needed o PNET ≠Carcinoid o Benefit of targeted agents rela@ve to PRRT, SSTa, chemotherapy, and/or liver-‐directed therapy unknown
o Op@mal sequence unknown o Eventual resistance to therapy is the rule
Benefits
Toxicity
Conclusions
NEXT STEPS
§ Optimize sequence § Explore mechanisms of resistance § “personalized” medicine