Pacheco, Richard R.

Pelvicalyceal invasion (PCI) is a relatively novel pT3a staging parameter for renal cell carcinoma (RCC) nephrectomies. While interobserver reproducibility staging studies of sinus/vascular invasion in RCC exist, a similar evaluation for PCI has not been performed. Moreover, in our experience, there is also diagnostic variability in how pathologists interpret PCI. Herein, we explore interobserver reproducibility among genitourinary (GU) pathologists. Twenty hematoxylin and eosin-stained digitized slides from RCCs (all grossly approaching the renal pelvis) were distributed to 31 GU pathologists to classify each as PCI or not PCI based on their respective clinical practices; slides with concomitant sinus/fat/vascular invasion were excluded. Slides were then evaluated for the following 4 morphologic features: tumor abutting renal pelvis, tumor pushing/indenting into the renal pelvis, polypoid configuration of tumor into the renal pelvis, and tumor eroding through renal pelvic urothelium. Interobserver reproducibility was assessed, and the morphologic features were correlated with PCI. Relationships between pathologists’ interpretations, morphologic features, and PCI were evaluated using hierarchical clustering. Although the diagnosis of PCI was relatively uniform with a majority agreement (>67%) reached in 16/20 slides, overall interobserver reproducibility was only moderate (kappa=0.601). While all 4 morphologic features were sensitive for PCI, polypoid configuration of the tumor into the renal pelvis and the tumor eroding through the renal pelvic urothelium were most specific (90%, 100%, respectively). Although we show general consensus among genitourinary pathologists on PCI assessment, clarifying the diagnostic guidelines with specific criteria should be included in pathologic staging systems.

AimsCompared to lymphovascular invasion (LVI), hilar soft tissue invasion (HSTI) is a relatively novel pT2 staging parameter in testicular germ cell tumours (GCT), associated with metastasis at the time of diagnosis and a risk factor for disease relapse. Given diagnostic variability in how pathologists interpret HSTI, herein we explore interobserver reproducibility among genitourinary (GU) pathologists.MethodsTwenty haematoxylin and eosin‐stained digitized slides from testicular GCTs were distributed to 30 GU pathologists to classify each as HSTI or not HSTI based on their respective clinical practices; slides with concomitant LVI were excluded. Slides were then evaluated for the following morphologic features: tumour invasion beyond prominent hilar vessels, tumour invasion beyond the rete, tumour invasion beyond the level parallel to hilar adipose tissue and tumour within (directly touching) hilar adipose tissue. Interobserver reproducibility was assessed, and the morphological features were correlated with HSTI. Relationships between pathologists' interpretations, morphological features and HSTI were evaluated using hierarchical clustering.ResultsAlthough the diagnosis of HSTI was relatively uniform with a majority agreement (>67%) reached in 17/20 slides, overall interobserver reproducibility was only moderate (kappa = 0.50). Morphological features such as invasion beyond the level parallel to hilar adipose tissue and invasion into fat were more sensitive and specific for HSTI, while features such as invasion beyond hilar vessels and invasion beyond rete were less useful.ConclusionsThese findings suggest that more specific defined diagnostic criteria for testicular GCT HSTI are needed.

Purpose: To investigate risk factors for recurrent or new tumor in patients with conjunctival melanoma. Methods: Retrospective review of patients with conjunctival melanoma managed on the Ocular Oncology Service, Wills Eye Hospital from 1974 to 2019. Results: There were 540 patients with mean follow-up of 57.6 months, of whom 176 (33%) had recurrent or new tumor formation. Risk factors for recurrent or new tumor on univariate analysis included presentation at older age (OR: 1.02 [1.01–1.03] per 1-year increase in age, p = 0.002), history of prior conjunctival surgery (OR: 1.62 [1.05–2.49], p = 0.03), worse visual acuity at presentation (OR: 1.76 [1.04–2.98] per 1 log-unit increase, p = 0.04), more advanced AJCC clinical T-subcategory (OR: 1.08 [1.02–1.14] per 1 subcategory increase, p = 0.01), tumor primary location in tarsal conjunctiva (OR: 1.80 [1.09–2.98], p = 0.02), and secondary tumor involvement of the fornix (OR: 1.68 [1.06–2.65], p = 0.03), and eyelid (OR: 1.92 [1.07–3.43], p = 0.03). Risk factors on multivariate analysis using all demographics, clinical features, and tumor location included presentation at older age (OR: 1.02 [1.00–1.03], p = 0.01), history of prior conjunctival surgery (OR: 1.84 [1.16–2.94], p = 0.01), and more advanced AJCC clinical T-subcategory (OR: 1.07 [1.01–1.13] per one subcategory increase, p = 0.03). Conclusion: On multivariate analysis, the strongest predictors of recurrent or new tumor formation following treatment of conjunctival melanoma included older age, history of prior conjunctival surgery, and advanced AJCC T-subcategory. These results suggest that earlier detection and the first surgery in conjunctival melanoma management are critical for prevention of recurrent or new tumor, and we recommend prompt referral to an experienced surgeon.