Diagnostic Performance of 18F-rhPSMA-7.3 PET in Men with Newly Diagnosed High-Risk Prostate Cancer and Negative Conventional Imaging

PURPOSE/OBJECTIVE(S)

The LIGHTHOUSE study (NCT04186819) evaluated high affinity radio hybrid (rh) prostate-specific membrane antigen (PSMA)-targeted PET radiopharmaceutical, 18F-rhPSMA-7.3, in patients across a range of prostate cancer (PCa) risk groups. Here, we report findings from a subgroup with high/very high-risk PCa who had no evidence of nodal or metastatic disease on conventional imaging.

MATERIALS/METHODS

Treatment-naïve patients scheduled for radical prostatectomy (RP) plus pelvic lymph node (PLN) dissection underwent PET 50-70 min after IV administration of 296 MBq 18F-rhPSMA-7.3. Local readers interpreted the scans prior to RP (performed ≤60 d post-PET) and ahead of a blinded read by 3 central readers. If the local read indicated M1 disease, verification (biopsy, surgery, or confirmatory follow-up imaging) of PET-positive M1 lesions was attempted before treatment. The present analysis evaluates the 18F-rhPSMA-7.3 sensitivity and specificity for detection of PLN metastases in all high/very high-risk patients with negative conventional imaging at baseline who underwent 18F-rhPSMA-7.3 PET and subsequent surgery. Histopathology was used as the standard of truth (SoT). Additionally, the M1 verified detection rate (VDR; % of patients with true positive (TP) M1 lesions using histopathology or follow-up imaging as SoT out of all patients scanned) was assessed in an extended population of all patients who had 18F-rhPSMA-7.3 PET regardless of surgery.

RESULTS

The sensitivity and specificity for PLN detection among 174 men with very/high-risk PCa and negative conventional imaging ranged from 24-33% and 92-96%, respectively, across readers (Table 1a). Across readers, M1 lesions were identified in 28-51 of the 197 patients in the extended population, giving an overall M1 detection rate of 14-26%. Of the identified lesions, 16-25 were successfully verified (predominantly using follow-up imaging as SoT) as TP, providing a M1 VDR of 8.1-13% (Table 1).

CONCLUSION

18F-rhPSMA-7.3 PET provides clinically useful information on the presence of both N1 and M1 disease prior to surgery in high/very high-risk PCa patients who were staged N0 M0 on conventional imaging. Specifically, up to 13% of such patients had verified TP M1 lesions. 18F-rhPSMA-7.3 may guide treatment, potentially helping such patients avoid ultimately futile surgery.

Development and Validation of an Objective Scoring Tool for Robot-Assisted Partial Nephrectomy: Scoring for Partial Nephrectomy

Objective: To develop a structured and objective scoring tool for assessment of robot-assisted partial nephrectomy (RAPN): Scoring for Partial Nephrectomy (SPaN). Materials and Methods: Content development: RAPN was deconstructed into 6 domains by a multi-institutional panel of 10 expert robotic surgeons. Performance on each domain was represented on a Likert scale of 1 to 5, with specific descriptions of anchors 1, 3, and 5. Content validation: The Delphi methodology was utilized to achieve consensus about the description of each anchor for each domain in terms of appropriateness of the skill assessed, objectiveness, clarity, and unambiguous wording. The content validity index (CVI) of ≥0.75 was set as cutoff for consensus. Reliability: 15 de-identified videos of RAPN were utilized to determine the inter-rater reliability using linearly weighted percent agreement, and Construct validation of SPaN was described in terms of median scores and odds ratios. Results: The expert panel reached consensus (CVI ≥0.75) after 2 rounds. Consensus was achieved for 36 (67%) statements in the first round and 18 (33%) after the second round. The final six-domain SPaN included Exposure of the kidney; Identification and dissection of the ureter and gonadal vessels; Dissection of the hilum; Tumor localization and exposure; Clamping and tumor resection; and Renorrhaphy. The linearly weighted percent agreement was >0.75 for all domains. There was no difference between median scores for any domain between attendings and trainees. Conclusion: Despite the lack of significant construct validity, SPaN is a structured, reliable, and procedure-specific tool that can objectively assesses technical proficiency for RAPN.

Transforming the Perioperative Treatment Paradigm in Non-Metastatic RCC-A Possible Path Forward

In 2017, there is no adjuvant systemic therapy proven to increase overall survival in non-metastatic renal cell carcinoma (RCC). The anti-PD-1 antibody nivolumab improves overall survival in metastatic treatment refractory RCC and is generally tolerable. Mouse solid tumor models have revealed a benefit with a short course of neoadjuvant PD-1 blockade compared to adjuvant therapy. Two ongoing phase 2 studies of perioperative nivolumab in RCC patients have shown preliminary feasibility and safety with no surgical delays or complications. The recently opened PROSPER RCC trial (A Phase 3 RandOmized Study Comparing PERioperative Nivolumab vs. Observation in Patients with Localized Renal Cell Carcinoma Undergoing Nephrectomy; EA8143) will examine if the addition of perioperative nivolumab to radical or partial nephrectomy can improve clinical outcomes in patients with high risk localized and locally advanced RCC. With the goal of increasing cure and recurrence-free survival (RFS) rates in non-metastatic RCC, we are executing a three-pronged, multidisciplinary approach of presurgical priming with nivolumab followed by resection and adjuvant PD-1 blockade. We plan to enroll 766 patients with clinical stage ≥T2 or node positive M0 RCC of any histology in this global, randomized, unblinded, phase 3 National Clinical Trials Network study. The investigational arm will receive two doses of nivolumab 240 mg IV prior to surgery followed by adjuvant nivolumab for 9 months. The control arm will undergo the current standard of care: surgical resection followed by observation. Patients are stratified by clinical T stage, node positivity, and histology. The trial is powered to detect a 14.4% absolute benefit in the primary endpoint of RFS from the ASSURE historical control of 55.8% to 70.2% at 5 years (HR = 0.70). The study is also powered to detect a significant overall survival benefit (HR 0.67). Key safety, feasibility, and quality of life endpoints are incorporated. PROSPER RCC exemplifies team science with a host of planned correlative work to investigate the impact of the baseline immune milieu and changes after neoadjuvant priming on clinical outcomes.

Genome-wide methylation profiling to identify potential epigenetic biomarkers associated with response to sunitinib in metastatic renal cell cancer (mRCC) patients (pts)

4566

Background: There exists a significant heterogeneity in the clinical response to sunitinib among pts treated for mRCC and, thus, a biomarker which would predict pts’ response up front would be an invaluable tool in the clinical management of these pts. In this regard, whole genome methylation array was performed between 2 extreme groups: sunitinib responders (RES) and non-responders (NRES) to identify differentially methylated genes between these subsets of pts. Methods: mRCC pts who received sunitinib therapy with available frozen nephrectomy tissues (stored at -80°C) and clinical data were identified. RES were identified as pts with progression free survival (PFS) of > or =11 months (mos) and NRES as those with PFS < or = 3 mos. After DNA extraction and quality assurance according to standard protocol, whole genome methylation array was performed using Infinium Humanmethylation450 BeadChip Kit - Illumina. Data normalization was achieved by subset-quantile within array normalization (PMID: 22703947). Differentially methylated regions were identified using logit transformed Beta values, using an F-test after shrinking variance via empirical Bayes (PMID: 21118553). Results: Total of 13 pts who received sunitinib therapy with available frozen nephrectomy tissues were identified. Of the 13, 5 pts qualified for each RES and NRES cohort as described in methods section. All pts in RES group had clear cell subtype. Two pts of the NRES group were of non-clear cell subtype. The QC plots showed that all arrays were successful. For RES vs. NRES, one genomic location, DENND2D, was significantly hypermethylated in the RES group with a false discovery rate (FDR) of <10%. DENND2D has been identified as a tumor suppressor-like gene in non-small cell lung cancer and melanoma cell lines in the recent past. Conclusions: In this study, DENND2D was significantly hypermethylated in sunitinib RES compared to NRES among mRCC pts. Data analysis with a less stringent FDR is also being pursued. Technical validation as well as clinical validation of DENND2D utilizing a larger pt cohort are ongoing.

Microperforations of surgical gloves in urology: minimally invasive versus open surgeries

INTRODUCTION

Surgical glove integrity is important in preventing wound infections and reducing patient mortality. Rates of perforations have been studied in many surgical subspecialties, but glove perforations specific to urology have not been investigated previously. This study aims to determine the incidence of glove perforations during urological surgeries and to investigate differences between open, laparoscopic, and endoscopic procedures.

MATERIALS AND METHODS

A total of 180 gloves were collected from various urological procedures performed at our institution: 59 from endoscopic, 72 from laparoscopic, and 49 from open cases. The gloves were tested for defects by both the water load test and electrical conductance testing. The frequency of defects for each type of procedure along with the length of wear, surgeon experience, and glove brand was analyzed.

RESULTS

Glove defects were detected in 29% of all cases. Microperforations encompassed the majority of the glove defects (23.3%) and were detected in 15.2%, 25.0 %, and 30.6% of the endoscopy, laparoscopic and open surgical cases, respectively. The frequency of perforations noted in the minimally invasive procedures was significantly different across the groups (p < 0.01). There was no statistical significant correlation between glove defects and operation time, surgeon experience, and glove brand.

CONCLUSIONS

The rates of glove perforation (29%) in urological procedures were higher than expected. Given the high rates of glove perforations found, double gloving in urological surgeries may offer a solution to the increased risk for cross contamination from microscopic perforations.

Clinical decision making using teleradiology in urology

OBJECTIVE

Using a personal computer-based teleradiology system, we compared accuracy, confidence, and diagnostic ability in the interpretation of digitized radiographs to determine if teleradiology-imported studies convey sufficient information to make relevant clinical decisions involving urology. Variables of diagnostic accuracy, confidence, image quality, interpretation, and the impact of clinical decisions made after viewing digitized radiographs were compared with those of original radiographs.

MATERIALS AND METHODS

We evaluated 956 radiographs that included 94 IV pyelograms, four voiding cystourethrograms, and two nephrostograms. The radiographs were digitized and transferred over an Ethernet network to a remote personal computer-based viewing station. The digitized images were viewed by urologists and graded according to confidence in making a diagnosis, image quality, diagnostic difficulty, clinical management based on the image itself, and brief patient history. The hard-copy radiographs were then interpreted immediately afterward, and diagnostic decisions were reassessed. All analog radiographs were reviewed by an attending radiologist.

RESULTS

Ninety-seven percent of the decisions made from the digitized radiographs did not change after reviewing conventional radiographs of the same case. When comparing the variables of clinical confidence, quality of the film on the teleradiology system versus analog films, and diagnostic difficulty, we found no statistical difference (p > .05) between the two techniques. Overall accuracy in interpreting the digitized images on the teleradiology system was 88% by urologists compared with that of the attending radiologist's interpretation of the analog radiographs. However, urologists detected findings on five (5%) analog radiographs that had been previously unreported by the radiologist.

CONCLUSION

Viewing radiographs transmitted to a personal computer-based viewing station is an appropriate means of reviewing films with sufficient quality on which to base clinical decisions. Our focus was whether decisions made after viewing the transmitted radiographs would change after viewing the hard-copy images of the same case. In 97% of the cases, the decision did not change. In those cases in which management was altered, recommendation of further imaging studies was the most common factor.