Discrimination of benign and malignant human tissue during percutaneous interventions using optical spectroscopy techniques

Although major advances are made in cancer imaging, daily practice in
diagnostic procedures and radiological interventions are still hampered by
inadequate recognition of tumor tissue at the time of the procedure. For
example, recent studies have reported varying figures of overall accuracy for
percutaneous lung biopsies, which respectively range between 67% and 96%. Also
for liver biopsies as well as liver RFA ablation, precise positioning of the
specific needle is essential for successful diagnosis or treatment. Although
positioning of the biopsy needle or RFA needle can be guided by ultrasound,
X-ray or CT scanning, visualization of relevant structures is often limited due
to thresholds in contrast and image resolution.
Here we propose to tackle these shortcomings by using optical spectroscopy
techniques (diffuse reflectance spectroscopy and fluorescence spectroscopy)
that allow precise real-time monitoring of tissue characteristics at the tip of
the biopsy needle or radiofrequency ablative probe. By illuminating specific
tissue with a selected light spectrum and subsequent analysis of the
characteristic scattering, absorption and luminescence patterns, it is possible
to obtain an *optical fingerprint* of the tissue and to discriminate between
benign and malignant tissue. In this way optical spectroscopy may be more
sensitive in tissue discrimination than conventional imaging techniques.
Incorporation of optical spectroscopy technology into current diagnostic or
therapeutic tools, e.g. in biopsy needles, could improve significantly the
accuracy of the intended procedure and thus clinical outcome.
Earlier, we have developed an optical spectroscopy system for measurement of
tissue characteristics. The concept has been tested on excised human tissue. We
were able to differentiate between normal tissue (including benign tumors) and
malignant tissue with a sensitivity and specificity of >95% within patient
analysis. Recently, these results have been confirmed by the *OpSpect* study
(NL32233.031.10), in which lung, liver and breast tissue were measured in vivo
during surgery. Comparison studies in the literature have demonstrated maximum
sensitivity and specificity percentages to be 83%. Our results are better
compared to literature, since we are able to detect tumor using the normal
tissue as internal reference for individual patients, hence removing
inter-patient variation.

We conclude from our results we are able to reliably detect - in vivo - the
difference between normal tissue and malignant tissue within an individual
patient by means of optical spectroscopy.
The aim of the present study is to implement and test these results in a
percutaneous clinical setting during standard percutaneous procedures. For this
purpose a specially designed *smart* biopsy needle was developed. The
radiologist will perform the core biopsy according to standard protocols. The
only difference will be that during insertion of the biopsy needle optical
measurements will be performed.

So, in conclusion, optical measurements will be performed in a clinical setting
during standard breast/lung/liver biopsy or prior to liver RFA. The results of
optical spectroscopy will be compared to histopathology results of the tissue
sample.

In our previous studies using optical spectroscopy, both ex vivo and in vivo,
most promising results were obtained for tissue diagnosis. Moreover, invasive
procedures for biopsy or radiofrequency ablation in these tissues are all very
common. Thus, the possibility of incorporation of optical spectroscopy in
needles for specific invasive procedures, such as biopsies or RFA ablations,
would be of significant additional clinical value in these tissues / organs.
For this purpose a specially designed optical biopsy needle was developed,
which can measure the tissue which is going to be sampled. With this
observational study we aim to prove that such a needle can be used in the
workflow of standard percutaneous interventions. Optical spectroscopy
measurement of the lesions will be compared to standard histopathological
analysis, as golden standard.

Primary Objective:
In this observational study we aim to evaluate whether optical spectroscopy can
correctly diagnose malignant tissue in the existing clinical workflow of
percutaneous interventions in breast, lung and liver.

Secondary Objective:
During the measurement procedure, possible improvements of the measurement
hardware will be recorded. Analysis of this documentation will provide
information for possible alterations of hardware design for improved clinical
applicability in the future. Special attention will be paid to observe how the
procedure fits in the standard workflow of the radiologist.

Further clinical development and perspectives:
This present proposed study fits within the project line *development of smart
optical devices*. It is the aim that we ultimately develop a smart optical
needle that can guide biopsies and radiofrequency procedures. This would allow
improvement of diagnosis as well as minimal invasive treatment by RFA.
Within this trajectory three consecutive steps are foreseen:
1) Ex vivo measurements for optimizing of the technique (accomplished)
2) In vivo measurements in controlled setting for further proof of principle
(accomplished: OpSpect Study, NL32233.031.10)
3) Implementation and testing in percutaneous clinical setting (present
proposal).

General
The study is designed as an observational study. Patients eligible for
inclusion into the study are patients who are already admitted to The
Netherlands Cancer Institute (NKI-AvL) for a regular percutaneous core biopsy
in breast, liver or lung or prior to a percutaneous liver RFA.
Measurements will only be performed in the interventional room prior to
standard core biopsy of a suspected malignancy in breast, lung or liver or
prior to planned RFA in liver metastases. Optical measurements will be
performed using a specially designed CE-marked optical biopsy needle. This
needle is able to perform optical measurements exact in the tissue that is
going to be sampled.

Organs specific issues
Targeted lesions in breast, lung and liver will involve any lesion that is
suspected to be malignant.

Procedures
It is important to note that biopsy procedures will be performed according to
the standard method of performing a core biopsy, except that during insertion
and just before tissue sampling, optical measurements will be performed.
The radiologist responsible for the procedure will identify the target lesion
by CT or ultrasound imaging as usual. The optical biopsy needle will be
inserted like a standard biopsy needle. During this insertion real-time optical
measurements will be performed. The anticipated total additional time for the
whole procedure will be less than 5 minutes for breast/long/liver biopsy
patients and maximal 10 minutes for patiens who will undergo percutaneous liver
RFA. Sampled tissue will be analysed by the pathologist via a standardized
protocol. Histopathological analysis of the tissue sample will be provided for
comparison with optical analysis.

Because of the nature of this test, we do not expect any adverse events to
occur that are related to technology of the optical
spectroscopy hardware. Collected data will not be provided to the physicians
and the planned procedure will not be influenced by the
optical measurements.