Circulating tumour cells by diagnostic leukapheresis mirror primary tumour heterogeneity in non-small cell lung cancer

Circulating tumour cells (CTCs) are a strong predictor of prognosis(1-3)
and can be used as a biomarker for early detection of systemic cancer spread,
therapy monitoring, and nowadays also for single cell genomics. Obtaining
relevant information from blood by so called *liquid biopsies* is thus a simple
method to detect tumour cells. Therapeutic decisions in lung cancer are
increasingly dependent on adequate tumor tissue biopsies. However, amongst
others, tumour heterogeneity and technical issues with the handling of tissue
allow adequate diagnosis in only part of patients. CTCs may help to bypass
these problems: CTCs do not have the issue of contamination with normal cells
and DNA/RNA from leukocytes that come with this technique can be harvested in
the same run(4).
Therefore, CTCs may replace current tumour biopsy practices when an adequate
numbers of tumour cells can be detected, while also giving the option for
further mutation analysis(5,6).
Immunomagnetic enrichment of cancer cells from blood samples expressing
membranous epithelial cell adhesion molecule (EpCAM) protein led to the
development of the FDA approved CellSearch system, nowadays the most widely
used standard for CTC detection. The relevant detection rate is set on >2 or >5
CTCs per 7.5 ml blood sample. Clinical use of CTCs is currently limited in
NSCLC because all systems fail to detect CTCs at an acceptable rate and at a
sufficient high yield (for mutation analysis) in a large fraction of patients.
For NSLC, CTCs are observed in 26 to 49% of patients with metastatic disease.
In nonmetastatic disease one CTC per 7.5 ml is observed in only 5 to 24% of
patients. Extrapolation of CTC frequency distribution in 7.5 ml of blood from
patients with metastatic breast, colon and prostate cancer showed that probably
all these patients had CTCs in circulation, but the sample volume was not
sufficient to detect them in all patients(7). A possible solution for this
problem would be to significantly increase the blood volume. This can be
achieved with leukapheresis that has been specified to increase CTC detection
by means of a filter (e.g. Vycap). We hope this would provide a more reliable
detection of CTCs at a higher frequency, and that by using this technique CTC*s
can be found in sufficient high yield, even in nonmetastatic disease. We will
study these issues within the European CANCER-ID consortium, a public-private
partnership supported by Europe*s Innovative Medicines Initiative (IMI) with
currently 38 partners aiming at clinical validation of blood borne biomarkers
and establishing standard protocols for these. Leukapheresis, a standard
clinical method to isolate mononuclear cells (MNCs) from blood, is currently
used as routine practice in hematological diseases. Usually one to five liters
of blood is processed in adults. Diagnostic leukapheresis (DLA) has previously
been studied in solid cancer patients. Median total processed blood volume for
lung cancer was 2,6 l (1,4 - 11,0). This resulted in 56 mL (40 - 156) volume of
DLA product with 40.108 MNCs. The detection rate of CTCs in peripheral blood
was 22% versus 56% in DLA. The procedure took one hour without adverse
events(8).
In this study we will first explore the frequency and number of CTCs in all
stages of NSCLC. Our hypothesis is that CTCs mirror the primary tumour
heterogeneity at different stages of disease. Therefore, we will combine
diagnostic leukapheresis with single cell genetics to study tumour
heterogeneity for the prediction of therapy response in different stages of
NSCLC patient groups.

Primary aim: Frequency of identifying CTCs and their numbers in apheresis
product before and after (during) treatment.
Secondary aim: Prediction of response to therapy in different
stages of NSCLC patient groups with the change in CTC numbers.

2.1. Study design:
The clinical value of diagnostic leukapheresis (DLA), a more invasive and more
time consuming procedure than a venapuncture, will be related to the molecular
and functional characterization of CTCs in the context of personalized
molecular therapies. Screening high volumes of blood will enable a true liquid
biopsy for patients with NSCLC and may open the possibility of using CTCs as
biomarkers to guide and monitor systemic therapies even in the adjuvant therapy
setting. Therefore, we want to study the enumeration of CTCs by DLA and
determine whether CTCs can be used to detect single cell DNA abarrations and
mutations to determine treatment strategies in four different clinical
situations.

2.2. Study groups:
1. Patients with advanced NSCLC with mutations before and 3 weeks after the
first treatment of any treatment line (n=20).
2. Patients with advanced NSCLC without detectable mutations before and 3 weeks
after the first treatment (n=20).
3. Patients before and after chemoradiotherapy 3 weeks after the first
treatment in stage III NSCLC (n=20).
4. Patients with resectable NSCLC before and after surgery (n=20).

Ad 1 to 4. In case no CTCs will be determined in the first 3 patients with 3 L
DLA, we will extend the apheresis time to extract 5 L blood. If this doesn*t
produce results in the next 5 patients then no new patients will be included in
that group.
Ad 4. CTCs will be collected before surgery from a peripheral venipuncture (7.5
mL blood), DLA will be collected just before surgery, during operation before
the clamps are placed on the pulmonary vein from the pulmonary vene coming from
the lobe harboring the malignancy and 14 days after surgery, before adjuvant
treatment is initiated.
Ad 1-4. Before and after DLA, a routine peripheral venipuncture for CTC (7.5 mL
blood) will be performed.
Ad1-4. If no CTC*s can be found in the first procedure, then no second
apharesis will be performed.

2.3. Statistics:
Assuming a prevalence of CTCs in apheresis products at baseline of 80% and 26%
(effect size 0,67) after or during treatment, in 20 patients we should detect
this difference at a significance level of 0.05 (two-sided) and with a power of
80%. Thus, we will study 40 diagnostic leukapheresis products (20 baseline and
20 during or after treatment).
For testing the hypothesis that the decrease in CTCs will determine a prolonged
survival, the change in CTCs (using either median changes, using logistic
regression or using a ROC curve(9)) will be associated with the
progression-free survival and overall survival using Kaplan-Meier survival
curves. Cox regression analysis will be performed with correction for
covariates such as disease stage, performance status and therapy.
Exploratory studies will be performed with descriptive statistics with DNA and
RNA blood-borne biomarkers.

During the investigation the patient will receive a moderate drain on their
energy, as the apharesis will take about one hour of time and two needles will
be inserted in their arms, which might be painfull or stressfull for the
patients. The pain is not severe however. Also we don't expect the
complications to occur that often, and the consequences of these complications
should be limited. Complications that can occur are strange sensations in
extremities, face of lips, bruises at the injection sites, or bleeding. An
infection of the insertion site is the only severe complication.