SPECTRA: Intraoperative Hyperspectral Imaging for Prostate Cancer
Trial statusRecruiting
Trial phaseNot applicable
Trial typeInterventional
Biological sexMale
Age18+
SponsorMassimo Valerio
About this trial
Radical prostatectomy (RP) is currently one of the gold standard of care in managing localized prostate cancer (Pca). However, despite advances in Robot-Assisted RP (RARP), cancer cells may still be found (up to 42%) at the edge of the removed prostate specimen, commonly referred as Positive Surgical Margins (PSM). PSM are well-known risk factors of adverse oncological outcomes (biochemical recurrence).
It is therefore essential to characterize intraoperatively the tumor extent. In RARP, the leading intraoperative margin assessment (IMA) technique, NeuroSAFE is based on Intra-operative frozen section (IFS) of both prostate lateral sides sampled at 3-5 mm intervals. However, despite proven oncological benefit, adoption is still limited to expert centers due to a prolonged operative time, loss of integrity of the surgical specimen for definitive histopathological analysis, omission of regions (apex, base), and logistical complexity (trained uropathologist, cryostat…).
With this prospective single-center project, we wish to explore the use of non-invasive optical imaging technology, HyperSpectral Imaging (HSI), as a reliable alternative for examination of the ex-vivo prostate specimen to
1. Distinguish normal versus tumor tissue at the surgical margin in fresh specimen. 2. Quantify whether HSI can be integrated into the intraoperative workflow (feasibility), thus, image acquisition time and inference time should then be evaluated.
Eligibility criteria
This trial does not accept healthy volunteersQualifiers
Histological diagnosis (biopsy) of Pca
Prostate volume < 100 cm3 (measured on preoperative MRI) for a smooth acquisition with the hyperspectral device
Capable of forming his own views
Disqualifiers
Unwilling to participate.
Inability to follow study procedures, for example due to language problems, psychological disorders, dementia, etc.
Trial design
Design model
Single group
Treatments tested in this trial
HyperSpectral imaging acquisition of ex-vivo prostate specimen prior formalin fixation (fresh acquisition)..
Procedure/SurgeryThe study specific specimen handling consists in a multi-Angle HSI Image Acquisition before routine pathology. To enhance the co-registration between the two modalities, a black-ink may be applied on the ex-vivo specimen based on a MRI suspicious region of extra-prostatic extension (EPE Grade ≥ 2). This procedure follows prior robust registration work between ex-vivo HSI and histology evaluation. The identification of this region will rely on the standard pre-operative MRI review already performed as part of routine care within our prostate cancer center before each radical prostatectomy.
Treatment groups
26 Participants
are divided into 1 treatment groupGroup A: Participants with prostate cancer undergoing robot-assisted prostatectomy. HyperSpectral imaging.Experimental treatment 1 intervention
Trial outcomes
Primary outcomes
1
Tumor extent
Dice similarity coefficient between HSI-derived and histopathology spatial overlap between the HSI tumor map and co-registered histopathology
Time frame
HSI acquisition on Day 0 (day of surgery); reference histopathological assessment on postoperative Days 7Secondary outcomes
1
Boundary delineation
Normalized surface distance between HSI-derived andhistopathological tumor contours. Contour agreement
Time frame
HSI acquisition on Day 0 (day of surgery); reference histopathological assessment on postoperative Days 72
Region overlap
Jaccard index between HSI-derived and histopathological tumor masks intersection over union
Time frame
HSI acquisition on Day 0 (day of surgery); reference histopathological assessment on postoperative Days 73
Foreground discrimination
Pixel-wise sensitivity of HSI for tumor classification. Proportion of histologically tumor-positive pixels classified as tumor
Time frame
HSI acquisition on Day 0 (day of surgery); reference histopathological assessment on postoperative Days 74
Background discrimination
Pixel-wise specificity of HSI for tumor classification . Proportion of histologically non-tumor pixels classified as non-tumor
Time frame
HSI acquisition on Day 0 (day of surgery); reference histopathological assessment on postoperative Days 7Other outcomes
1
Surgical margins
Diagnostic accuracy of HSI for detection of histologically positive margins
Time frame
HSI acquisition on Day 0 (day of surgery); reference histopathological assessment on postoperative Days 7Sponsors and contacts
Click on the lead sponsor to view all of their trials.
Massimo Valerio
Lead sponsor
University Hospital, Geneva
Sponsor institution
Source ClinicalTrials.gov