Transcript ASCO_2008_files/Berlin Non-CRC poster disc ASCO 2008

Poster Discussion
Jordan Berlin, M.D.
Ingram Associate Professor,
Medicine
Disclosures
• Yes, I have a Consultant or
Advisory Role to disclose.
–
–
–
–
–
–
–
Amgen
BMS
Genentech
Imclone
Immunogen
Pfizer
Sanofi
• Yes, I have Honoraria to
disclose.
– BMS
– Genentech
– Sanofi-Aventis
• DSMB
– Pfizer
• Took a pen
– Genentech, but I promise
I’ll give it back
Adjuvant chemotherapy alone vs
chemoradiation after curative
resection for pancreatic cancer:
Feasibility results of a randomized
EORTC/FFCD/GERCOR phase II/III
study (40013/22012/9203)
1JL.
Van Laethem, 2E. Van Cutsem, 3P.
Hammel, 4F. Mornex, 5D. Azria, 6G. Van
Tienhoven, 7M. Peeters, 8M. Praet, 9V.
Budach, 2K. Haustermans
1Hôpital
Universitaire Erasme, Brussels (BE) - 2U.Z. Gasthuisberg,
Leuven (BE) - 3Hôpital Beaujon, Clichy (FR) - 4CHU Lyon (FR) 5CRLC Val d’Aurelle, Montpellier (FR) - 6AMC, Amsterdam (NL) 7Universiteit Gent (BE) - 8EORTC, Brussels (BE) - 9Charite Berlin
(DE)
• Study design
– Multicentre phase II/III study
Arm A (n = 45):
Gemcitabine
x 4 cycles (4 week cycles)
Within 8 weeks of
surgery
R0 Margin status
R
Arm B (n = 45):
Gemcitabine x 2 cycles then
gemcitabine + XRT
Results
• Dose delivered was good
–Had 20-37% with dose reductions at
some point
• Compliance was pretty good with
chemo, but 11/45 did not receive XRT
and 4 more are unknown
• Toxicity was manageable
• DFS was 10.9 months for the entire
group
Discussion
• 300 mg/m2 gemcitabine with XRT was
possible in a multiinstitutional study
• A significant number of patients (>20%)
did not receive XRT
• DFS was short (~3 months shorter than
CONKO-01 gemcitabine arm)
– Course too short?
– XRT arm received even less treatment?
– Small study?
– Patient selection?
» >70% + for perineural invasion, for example
Adjuvant Therapy of Pancreas
Cancer
•42*
*Answer to the ultimate question according to the
5 books of the Douglas Adams classic Hitchiker’s
Guide to the Galaxy trilogy
EORTC set out to answer the ultimate
question
Adjuvant Pancreas Cancer
Discussion
• After decades with limited studies,
we know little more about adjuvant
therapy of pancreas cancer than we
did before they invented the CT scan
machine
• Chemotherapy appears to have
benefit (ESPAC-1 and CONKO-01)
The Ultimate Question
• We have no studies that either prove
or disprove the use of radiation in
adjuvant pancreas cancer
–GITSG
–EORTC
–ESPAC-1
–RTOG 9704
Localized Therapy
• Local control is an issue in the
adjuvant therapy of pancreas cancer,
–But we still have little, if any, systemic
control
–In the best studies, ~80% will be dead,
largely due to pancreas cancer, at 5
years
–Should we test a local control issue and
how do we best test it in a disease
where systemic control is so poor
Adjuvant Pancreas Cancer
• Are there alternatives?
–With creative design and endpoints, it
may be possible to test new agents in
the adjuvant setting
» After all, resectable pancreas cancer really
represents in most patients an earlier stage
of metastatic disease
–It is definitely time for a new paradigm
and approach to this disease
LV5FU2-cisplatin followed by
gemcitabine or the reverse
sequence in metastatic pancreatic
cancer: Preliminary results of a
randomized phase III trial (FFCD
0301).
E. Mitry, L. Dahan, M. Ychou, J. Arthaud,
M. Gasmi, J. Raoul, C. Mariette, J. M.
Phelip, L. Bedenne, J.F. Seitz, Fédération
Francophone de Cancérologie Digestive
• Study design
– Multicentre phase III study
– To compare the best sequence of chemotherapy
Arm A:
LV5FU2-P followed by Gemcitabine
at progression or toxicity
Stratification:
- Centre
- WHO PS (0,1 vs 2)
- Location (head vs other)
- GEM infusion rate
R
(Minimization)
Arm B:
Gemcitabine followed by LV5FU2-P
at progression or toxicity
Gemcitabine: 1000 mg/m² on D1 as a 30 mn infusion or with an infusion rate of 10
mg/m2/min* , weekly 7 weeks /8, then 3 weeks / 4
LV5FU2-P: 2-hour infusion of LV 200 mg/m2 followed by a FU bolus 400 mg/m2 and
46-hour infusion 2,400 mg/m2 every 2 weeks, with cisplatin 50 mg/m2 as a 2-hour
infusion on D1 or D2
(* Each participating centre had to always use the same administration method)
Results
• Response rates similar
–18.6% (LVU2-P 1st) vs 22% (gem 1st)
• PFS similar
–3.83 mos (LVU2-P 1st) vs 4.73 mos (gem
1st), p = 0.9
• OS similar
–6.63 mos (LVU2-P 1st) vs 8.03 mos (gem
1st), p = 0.77
What does this study tell us?
• Too small to be a true equivalency
trial
• LVFU2-P is not a replacement for
gemcitabine 1st line
• It is okay to use something without
gemcitabine 1st line on clinical trial
2nd line pancreas cancer
A PHASE 2 STUDY OF BEVACIZUMAB PLUS
ERLOTINIB IN PATIENTS WITH GEMCITABINEREFRACTORY METASTATIC PANCREATIC CANCER
AH Ko, E Dito, B Schillinger, AP Venook, EK
Bergsland, WM Korn, MA Tempero
A Phase II Trial of Sunitinib (S) in PreviouslyTreated Pancreas Adenocarcinoma (PAC),
CALGB 80603
E.M. O’Reilly, D. Niedzwiecki, D. Hollis, T. BekaiiSaab, T. Pluard, A. Duffy, F. Overcash, P. Ivy,
R.M. Goldberg
Results
• Ko, et al: erlotinib + bevacizumab (n =
32)
–1 PR (3.6%), 7 SD (25%)
–TTP 40 days (1.33 months)
–OS 104 days (3.5 months)
• O’Reilly, et al: sunitinib (n = 77)
–1 CR (1%), 14 SD (20%)
–PFS 1.35 months
–OS 3.2 months
Ko and O’Reilly Papers
• Both regimens had limited antitumor effect
– But both had some effect
• Why were they ineffective?
– Were these the right agents to hit the targets?
– Should these have been tested in first-line?
– Is VEGF and/or angiogenesis a relevant target
in pancreas cancer?
– Is targeting EGFR smart in a disease where
90% of patients have activating mutations?
VEGF/Angiogenesis as Target
• Pancreas cancers frequently overexpress VEGF,
PDGF, and PDGFR
• In tumor samples from patients with pancreatic
cancer, in some studies, ↑ VEGF expression
correlated with
– Increased microvascular density
– Increased incidence of liver metastasis
– Decreased survival
• Inhibition of VEGF or its receptors prevents
growth of pancreatic tumor xenografts in animal
models
• In other studies, microvessel density did not correlate
with prognosis (Ellis, et al Eur J Cancer 34:337-40, 1998)
2nd line pancreas cancer
Commentary
• While the drugs may not have worked
because the target was wrong
–Maybe the drugs did not hit enough
targets
–Maybe the drugs did not hit the right
combination of targets
–Maybe it is time to go back to the basics
Pancreas Cancer Overall
• In each setting, one theme emerges
– We need to rethink our study designs and
paradigms
• We have spent a decade looking for
incremental differences
– We achieved this goal with erlotinib and many
are displeased with that
– Let’s look for real change again
• Trials should be designed with intelligence
– Let science guide the trial-don’t just use it to
rationalize a foregone conclusion
Pancreas Cancer overall
• Like most disease sites we do our
trials empirically
–The science is used as a rationalization
• The flaws here are many
–Science should guide our choices rather
than be used to support what we already
decided to do
–Published lab data is highly biased
towards positive trials
» We need get the negative data as well.
• Editors, authors and reviewers take note
Long Term Outcome of Peptide Receptor Radionuclide Therapy in
441 Patients with Progressive Neuroendocrine Tumors Using
Yttrium-90 and Lutetium-177 labelled Somatostatin Receptor
Targeting Peptides
Dieter Hörsch, Vikas Prasad, Richard P. Baum
Department of Internal Medicine and Nuclear Medicine /
Center for PET/CT Zentralklinik Bad Berka
Is this the future of
radiation for NE tumors?
BAD BERKA PROCEDURE FOR PRRT
Studies before therapy
Renal scintigraphy
[99mTc- MAG3]
GFR measurement
[99Tc- DTPA]
90Y
/ 177Lu-DOTA-TATE Peptide Receptor Radiotherapy
Receptor PET/CT*
[68Ga-DOTA-NOC]
Infusion of aminoacid solution (- 0.5 til 4 hrs)
- 2 days
0
Infusion (15 min.) of
/ 177Lu- DOTA-TATE 1
90Y
2
3 days
Studies under therapy
177Lu-
DOTA-TATE WB scan
[planar scans for dosimetry]
177Lu-
DOTA-TATE- SPECT
of the tumor region
Blood sampling
Urine sampling
* Since July 2004. Previously, Tc-99m EDDA Hynic TOC (planar & SPECT) was performed.
In selected patients, also F-18 FDG and / or F-18 fluoride PET/CT is performed as well as MRI of the liver / bones
RESPONSE TO PRRT – DIFFERENT TREATMENT REGIMENS
Highest response rate was observed with Y-90 followed by
Lu-177 + Y-90 and Lu-177 DOTA-TATE alone
65
Percentage of Patients
61
54
Y-90
35 37
Y-90
Lu-177
Lu-177
31
Y-90
Lu-177
Y-90
Lu-177
CR/PR/MR
SD
8
9
Y-90
Y-90
Lu-177
PD
Horsch, et al Commentary
•
This study was conducted with:
– Highly selected patients with high
somatostatin receptor density of
primary tumor / metastases by 68GaDOTA-NOC receptor PET/CT
– All were “progressive”
– Not all patients treated were presented
» “Only progressive GEP NET patients
treated with at least 3 cycles of PRRT were
included in this analysis”
Horsch, et al Commentary
• Methods
–3 different strategies employed
» PRRT using Y-90,
» PRRT using Lu-177
» Combined use of Y-90 & Lu-177 labeled
DOTA-TATE
–It appears that they have been
modifying methods to reduce toxicity all
along
» And frequency of re-treatments
So, where is PRRT?
• There remain more questions than
answers (3 slides on the poster)
–Despite large numbers of patients
treated on these studies
» We don’t know best way to give
» Can we prevent nephrotoxicity?
• How bad is it?
» How bad is the heme toxicity?
• 7/454 (1.5%) developed MDS or AML, and not everybody may have been
followed long enough
» What about the other patients?
» What is the denominator (number of
patients rejected for this treatment)?
Should PRRT be moved forward?
• Certainly, it appears effective
– It also appears toxic
• What role this therapy will play needs to be
seen
– It needs multiinstitutional assessment before it
can even be considered for use off clinical
trials
» Who can give this?
» Who can do the studies needed for eligibility?
– Targeted (misnomer) therapies are showing
efficacy
» Will they make PRRT irrelevant?