Feasibility study of biomarker development for response prediction by large scale DNA mutational analysis of metastatic lesions.

Current chemotherapeutic regimens demonstrate response rates that could be
improved.
Hence, patients are exposed to (potential) harmful chemotherapeutic agents
without knowing whether or not they will respond to and benefit from treatment.
One of the focuses of molecular genetic research is to find biomarkers
predicting response to treatment and next-generation DNA sequencing is a highly
promising technology for this purpose. Genetic discrepancies can be seen
between metastases and the primary tumor. Therefore it is necessary to study
the mutational profile of safely and easily accessible metastatic lesions.

With this feasibility study we specifically aim to explore whether it is
possible to develop a predictive biomarker for chemotherapy response based on
next-generation DNA sequencing data, under ideal circumstances. For this, we
focus on a setting with 1. Relatively low heterogeneity in molecular pathways
leading to chemotherapy response (i.e. a monotherapeutic regimen: standard of
care treatment with irinotecan for metastatic colorectal cancer or other solid
tumors); 2. A primary endpoint most likely to be associated with the measured
genetic profile (i.e. short-term treatment response of the metastasis used for
DNA sequencing); and 3. Maximal contrast (i.e. comparing the best with the
worst responders).

Colorectal cancer is the third most common adult malignancy worldwide,
Approximately 20% of patients with colorectal cancer present with distant
metastasis (stage IV) at time of diagnosis, with a 5-year survival rate of
approximately 10%. Significant progress has been made in the treatment of
metastatic colorectal cancer, resulting in prolonged median survival from five
months to approximately two years. Irinotecan monotherapy has been widely
accepted as standard of care second line treatment with response rates of only
10-17%. No predictive biomarkers for irinotecan response have been found yet,
although response may correlate to plasma levels of the active metabolite
(SN-38) or carboxylesterase activity within the tumor.

For the proposed study we have chosen to select patients with metastases from
colorectal cancer, who have failed first-line palliative treatment and are
irinotecan naïve. This is explicitly not a drug study: patients will receive
standard of care second-line treatment with irinotecan monotherapy. Prior to
intervention with irinotecan patients will be subjected to a histological
biopsy of a metastatic lesion for DNA sequencing and analysis of
carboxylesterase. In case of definitive discontinuation of irinotecan (at any
time point for any reason) patients will again be subjected to a histological
biopsy. Additional analyses consist of blood samples for pharmacogenetic
research, pharmacokinetic analysis of circulating SN-38 (the active metabolite)
during first administration of irinotecan.

Primary objective:
To determine the feasibility of developing predictive biomarkers by large scale
DNA-sequencing for the response to irinotecan monotherapy as standard of care
treatment for metastatic colorectal cancer or other solid tumors.

Secondary objectives:
- To explore the correlation between survival and mutational profile of the
index lesion.
- To relate pharmacogenetics to the primary objective of the study.
- To determine changes in mutational profile of metastasis under the influence
of chemotherapy.
- To aid the development of strategies for statistical analysis by generation
of valuable human data.

Tertiary objectives:
- To correlate plasma pharmacokinetics of the active irinotecan metabolite
(SN-38) to the efficacy of irinotecan treatment.
- To determine carboxylesterase activity for irinotecan response apart from the
mutational profile and pharmacokinetics.
- To determine the safety of study related procedures.

Observational prospective multicenter biomarker discovery cohort study.

Burden in time for the individual patient:
- baseline screening: approximately 3 hours
- biopsy: 2 x 3 hours (biopsy itself approximately 15-30 minutes, afterwards
observation for at least 2 hours)
- pharmacokinetics: 24 hours (with an optional hospital admission overnight)

Burden of invasive procedures:
- 2 biopsies of a metastasis
- 13 (or 21; for Rotterdam cohort of patients) x 5 ml additional blood draws: 1
x 5 ml + 1 x 10 ml for pharmacogenetic analysis, 11 x 5 ml for pharmacokinetic
analysis (cycle 1: day 1 10 blood draws and day 2 1 blood draw with an optional
hospital admission overnight). The 11 additional blood samples will be taken
from a placed intravenous access

Risks: there is a small chance on complicatons (specified in Appendix C):
- biopsy: pain, bleeding, infection, allergic reaction
- blood draws: pain, hematoma, infection