Annual Meeting Proceedings Part 1 - American Society of Clinical ...

228s Gastrointestinal (Colorectal) Cancer
3601 General Poster Session (Board #34F), Mon, 8:00 AM-12:00 PM
Cost-effectiveness analysis per life-year gained based on predictors of
response for first-line metastatic colorectal cancer therapy in Spain.
Presenting Author: Cristobal Belda-Iniesta, Centro Integral Oncológico
Clara Campal (CIOCC), Madrid, Spain
Background: Economic crisis, increasing prevalence and cost of mCRC
management endanger availability of biological therapies for public health
systems due to budget limitations. We performed a cost-effectiveness
analysis based on markers of response/resistance including biological
therapies available in Spain for mCRC. Methods: We aimed to calculate
incremental cost-effectiveness ratio (ICER) per life-year gained (LYG) and
progression-free year gained based on predictive markers for mCRC.
Efficacy data include randomized trials (RT) that guided on-label uses of
bevacizumab and cetuximab. Control arms from these trials were used as
reference to calculate ICERs. Markers of clinical benefit (biological and
radiological) were included in this model. Toxicity as predictor of efficacy
was excluded for any therapy. Prices for drugs in Spain were assumed to
represent the best-value for each drug including all possibilities to reduce
pharmacy costs. For 1st line, median duration of therapy reported by RT
was used to calculate the final budget. 70kg and 1.7 m were used as
reference for patients dose calculations. If accessible, HR for PFS and OS
were used instead of medians. Results: K-Ras status and early response
measured by CT at 8 weeks were used as predictors of resistance and
increased efficacy for cetuximab-based combinations. We have not identified
any predictor marker for other drugs from RT. In this regard,
FOLFIRI�cetuximab combination obtained an ICER below the widelyproposed
Spanish threshold of 30,000 € per LYG if patients harbored wt
K-Ras tumors and evidenced an objective response at 8 weeks. Other ICERs
for different schedules were too distant from this limit, they will be
presented at the meeting. Multivariate analysis confirmed the robustness of
results. Conclusions: 1st line FOLFIRI�cetuximab therapy for wt K-Ras
patients that get an objective response measured by CT at 8 weeks is the
only cost-effective therapy option for mCRC below usual health-economic
thresholds for Spain. Our results are critical to design cost-effectiveness
based clinical guidelines for mCRC that will contribute to financial
sustainability of public health system in Spain.
3603 General Poster Session (Board #34H), Mon, 8:00 AM-12:00 PM
Improved early prediction of individual prognosis for patients with mCRC:
Joint modeling of tumor shrinkage with volume data for PFS and OS.
Presenting Author: Ulrich Robert Mansmann, Institute of Medical Informatics,
Biometry and Epidemiology, Klinikum Grosshadern, University of
Munich, Munich, Germany
Background: Piessevaux et al. (ESMO 2010) proposed a decrease in
RECIST-based tumor size of 20% at week 8 following 1st-line therapy for
mCRC as a predictor of clinical outcome (PFS and OS). We expanded upon
this idea (ASCO GI 2012). A tumor volume algorithm was developed
providing a better approximation to true tumor volume by using both the
longest and the longest orthogonal diameters of a target lesion. In this study
we compare the quality of early prediction-based individual patient
prognosis based on tumor change according to either RECIST or the tumor
volume algorithm. Methods: The prognostic method is combined with the
volume algorithm and applied to the data from 2 studies (OPUS, n�337;
CRYSTAL, n�1198) and 4 treatment regimens (FOLFOX4 �/- cetuximab
and FOLFIRI �/- cetuximab), in patients with mCRC. The influence of the
treatment regimen on early PFS and OS prognosis is studied by joint
modeling. The quality of early prognosis depending on the tumor size
assessment used is compared by the logarithmic scoring rule. Results:
Individual volume shrinkage and baseline volume are considered prognostic
factors. Individual predictions depend on the chemotherapy administered
and whether cetuximab is added. Equivalent tumor baseline volumes
and early volume changes, predict an up to 20% higher 1 year (y) PFS and
2y OS rate for FOLFIRI than for FOLFOX. The addition of cetuximab to
standard chemotherapy (CT) translates into a mean increase in shrinkage of
10% resulting in an improvement in 1y PFS rates of 23% and in 2y OS
rates of 18%. The quality of individual prediction over the 4 regimens is
combined in one logarithmic (log) score. The log score for the RECISTbased
prediction is 2.45 which is significantly higher than that for the
volume-based prediction of 1.97 (p�0.0001). Lower scores correspond to
a better prediction. Conclusions: The tumor volume algorithm enables a
more precise prediction of individual patient PFS and OS than RECISTbased
tumor assessments. Based on early tumor changes for CT, �/cetuximab,
it is possible to calculate quantitative information on a patient’s
prognosis. The model has high potential to guide individual clinical
decision making for mCRC patients.
3602 General Poster Session (Board #34G), Mon, 8:00 AM-12:00 PM
Aflibercept versus placebo in combination with FOLFIRI in previously
treated metastatic colorectal cancer (mCRC): Mean overall survival (OS)
estimation from a phase III trial (VELOUR). Presenting Author: Florence
Joulain, sanofi, Chilly Mazarin, France
Background: VELOUR evaluated the efficacy and safety of the novel fusion
protein aflibercept (VEGF Trap) in combination with FOLFIRI in mCRC
patients previously treated with oxaliplatin. Median OS showed significant
improvement with 12.06 months in placebo arm vs 13.50 months in
aflibercept arm (p�0.0032; HR�0.82 [95.34% CI: 0.71 to 0.94]) [Van
Cutsem, 2011]. Since survival data in oncology are usually right skewed,
median survival is preferred for regulatory purposes. However, mean
survival estimation can render a more meaningful estimate for long term
benefit of interventions, and be effectively applied to clinical and economic
decision support. Estimating mean OS also allows payers to derive an
estimation of total costs and outcomes in the population. The purpose of
this study was to estimate the mean OS for VELOUR trial. Methods: During
the trial follow-up period, mean OS is not observable and can be estimated
by extrapolating the trial Kaplan-Meier curve using a survival function.
Several standard parametric distributions were tested: exponential, weibull,
lognormal, loglogistic and gompertz. Akaike’s Information Criteria (AIC),
Bayesian Information Criteria (BIC) and graphical method were used to
evaluate the goodness of fit of the distributions. Models were run by
treatment arm separately or combining the two arms and using treatment as
covariate to control for variation. Results: Using AIC/BIC and graphical
method, loglogistic function best fit the VELOUR data both with and
without treatment as covariate. Weibull distribution is used for sensitivity
analysis. Conclusions: Using loglogistic function, mean OS benefit for
aflibercept in combination with FOLFIRI is at least 2.9 months (versus 1.4
months difference in median survival). The results have important implications
for clinical and economic decision support. Study NCT00561470
was funded by sanofi, in partnership with Regeneron.
Mean OS (months)
Placebo* Aflibercept* Difference
Loglogistic
By treatment separately 18.9 25.5 6.6
With treatment as covariate 20.3 23.2 2.9
Weibull
By treatment separately 14.9 17.9 3.0
With treatment as covariate 15.0 17.7 2.7
* � FOLFIRI
3604 General Poster Session (Board #35A), Mon, 8:00 AM-12:00 PM
Integrin genetic variants and stage-specific tumor recurrence in patients
with stage II and III colon cancer. Presenting Author: Pierre Oliver Bohanes,
University of Southern California Norris Comprehensive Cancer Center, Los
Angeles, CA
Background: Integrins are key elements in cancer biology regulating tumor
growth, angiogenesis and lymphangiogenesis through interactions of the
tumor cells with the microenvironment. Recent evidence showing that
integrins are critical in cancer dormancy suggests that their differential
expression or activity may be responsible for tumor recurrence. Moving
from the hypothesis that integrins could have different effects in stage II
and III colon cancer, we tested as a primary endpoint whether a comprehensive
panel of germline single nucleotide polymorphisms (SNPs) in integrin
genes could predict stage-specific time to tumor recurrence (TTR) in stage
II and III colon cancer patients. Methods: A total of 234 patients, 105
high-risk stage II and 129 stage III, treated with 5-fluorouracil-based
chemotherapy at the University of Southern California were included in this
study. The median follow-up time was 4.4 years. Whole blood samples were
analyzed for 22 germline SNPs in integrin genes using PCR-RFLP or direct
DNA-sequencing. Results: In the multivariate analysis,stage II colon cancer
patients with at least one G allele for ITGB3 rs4642 had higher risk of
recurrence (HR�4.027, 95%CI 1.556-10.421, p�0.004). This association
was also significant in the combined stage II-III cohort (HR�1.975,
HR 95%CI 1.194-3.269, p�0.008). The predominant role of ITGB3
rs4642 in stage II diseases was confirmed using recursive partitioning,
showing that ITGB3 rs4642 was the most important factor in stage II
diseases. In contrast, in stage III diseases, both ITGB1 rs2298141
(HR�1.909, 95%CI 1.054-3.459, p�0.033) and ITGA4 rs7562325
(HR�0.227, 95%CI 0.064-0.804, p�0.022) were associated with TTR.
The latter showed a significant interaction between stages (p�0.048).
Conclusions: This study identifies germline polymorphisms in integrin
genes as independent stage-specific prognostic markers for stage II and III
colon cancer. These data strengthen the role of tumor dormancy in early
colon cancer and may help to select subgroups of patients who may benefit
from integrin-targeted treatments.
Visit abstract.asco.org and search by abstract for the full list of abstract authors and their disclosure information.