FISH Analysis in Urothelial Cancer Michael Neat, Dr M Mason and Dr A Chandra Interphase FISH in urothelial carcinoma пЃ® Low sensitivity of urine cytology esp in low grade lesions вЂ“ Need for additional tests for detection and monitoring вЂ“ In conjunction with not in lieu of routine procedures пЃ® The UroVysion FISH assay вЂ“ First published 2000; 4 loci with best combined sensitivity from 10 candidates вЂ“ 2 FDA trials п‚§ 2001 FDA approved for detection of recurrence п‚§ 2005 вЂ“ вЂ“ вЂ“ вЂ“ вЂ“ вЂ“ Pts with haematuria No prev Hx Ca bladder Ca bladder histologically Dx in 50/497 (10.2%) FISH detected 69% of these, cytology 38% When TaG1 tumours excluded; FISH 83%, cytology 50% FDA approval 2005 pts with haematuria The FISH assay Abbott Molecular Mix of 4 probes labelled with 4 different fluorochromes Unstained ThinPrep slides Analysis/scoring criteria пЃ® Initially select morphologically abnormal cells вЂ“ вЂ“ вЂ“ вЂ“ Large nuclear size/irregular shape Patchy DAPI stain Cell clusters (non-overlapping) If no morphologically abnormal cells present, scan all cells Minimum analysis of 25 cells пЃ® FISH positive if: пЃ® вЂ“ в‰Ґ4 cells showing gain of at least 2 of #3, #7 & #17 вЂ“ в‰Ґ12 cells showing homozygous deletion of p16 i.e. no p16 signals Examples of abnormal signal patterns Increased copy no. of #3, #7 & #17 Homozygous deletion of p16 ICN & homozygous deletion of p16 Success rate пЃ® Analysis successful 58/59 (98%) cases вЂ“ 1/59 вЂ“ post treatment, probe hyb failed - ? DNA degraded вЂ“ 14/58 (24%) FISH positive пЃ® Highly reproducible assay when samples adequate вЂ“ 12/71 (17%) samples received insufficient material вЂ“ Caraway et al 65/1006 (6%) insufficient (cytospin) Performance of the assay пЃ® Halling & Kipp Eur Ren Genotourinary Dis. 2006;2:51-54 вЂ“ Mean sensitivity of FISH cf. cytology in 12 studies вЂ“ Cytology specificity higher than FISH (93% vs. 85%) пЃ® Stage/grade FISH (%) Cytology (%) Ta 67 28 Tis 97 73 T1 90 67 T2-T4 92 74 Grade 1 50 18 Grade 2 75 45 Grade 3 90 69 Stage/grade FISH (%) Cytology (%) All 72 (69-75) 42 (38-45) Excluding Ta 86 (82-89) 61 (56-66) Hajdinjak Urol Oncol. 2008;26:646-651 вЂ“ Meta-analysis (2477 FISH tests in 14 studies, cytology from 12) вЂ“ Cytology specificity higher than FISH (96% vs. 83%) пЃ® Conflicting data вЂ“ May et al. Urology 2007;70(3):449-53 п‚§ Conventional cytology can be better than FISH in experienced hands п‚§ Sensitivity 71% vs. 53.2% п‚§ Specificity 83% vs. 74% вЂ“ Moonen et al. Eur Urol 2007;51(5):1275-80 п‚§ No improvement over cytology in detection of recurrence п‚§ Sensitivity 40.6% vs. 39.1% п‚§ Specificity 89.7% vs. 89.7% Clinical applications пЃ® пЃ® пЃ® пЃ® пЃ® Detection of recurrence Gross or microscopic haematuria Anticipatory positive results вЂ“ FISH can detect tumour before clinically detectable by cytoscopy or cytology Helpful for clarifying equivocal cytology in patients with equivocal or negative cytoscopy ? detection of non-UC bladder tumours вЂ“ Histological variants detected on FFPEвЂ™s вЂ“ ? Exfoliating tumours Clinical applications (cont.) пЃ® пЃ® Follow-up post intravesical therapy вЂ“ BCG-associated inflammation makes cytoscopic & cytologic interpretation difficult вЂ“ Savic et al п‚§ 68 pts; NMIBC, post BCG п‚§ Both positive cytology and positive FISH predict failure of BCG п‚§ FISH superior when cytology non-definitive, i.e. equivocal, mild or moderate atypia вЂ“ Whitson et al п‚§ Positive FISH after IVT significant predictor of recurrence in multivariate analysis Detection of upper tract UC Author FISH (%) Cytology (%) Marin-Aguilera et al 76.7 36 Akkad et al 87.5 60 Disadvantages пЃ® пЃ® Cost Technical & interpretive difficulties вЂ“ Training вЂ“ equipment пЃ® False positives вЂ“ BK polyoma virus (rare) вЂ“ Tetraploidy п‚§ Reactive urothelial cells п‚§ Cells in S or G2 phase п‚§ ? Less specific predictor of malignancy пЃ® False negatives вЂ“ low-grade neoplasms if representative cells are not shed into the urine sample вЂ“ Lack of atypical cells on the slide used for FISH Conclusions Useful adjunctive assay to increase sensitivity in targeted patient populations пЃ® In routine use in many countries пЃ® Developing assay пЃ® Does earlier detection translate into decreased mortality? пЃ® Is negative predictive value sufficient to decrease the need for or frequency of cytoscopic follow-up? пЃ® Total FISH tests undertaken Total tests = 56 13 (41 patients) Negative Positive 43 Follow up data 41 45 40 35 30 25 20 15 10 5 0 lo gy 15 N o hi st o to lo gy H is pe rfo rm ed os co py al p at ie nt s 22 Cy st To t 37 Total number Cytology- Cytology- FISHHistology FISH Histology Correlation 18 11 0 5 10 4 12 4 15 Match Mis-match 4 20 25 Sensitivity and Specificity of FISH with histology FISH NEGATIVE HISTOLOGY NEGATIVE HISTOLOGY POSITIVE FISH POSITIVE 14 2 2 4 SENSITIVITY 66.7% SPECIFICITY 87.5% Cytology categories пЃ® пЃ® C1: Unsuitable for diagnosis C2: Benign пЃ® C3: Atypia, probably reactive (expected outcome вЂ“ 10-15% malignant) пЃ® C4: Atypia, probably malignant C5: Malignant пЃ® Sensitivity and Specificity of cytology with histology (C3=NEGATIVE) CYTOLOGY NEGATIVE (C 1,2,3) HISTOLOGY NEGATIVE HISTOLOGY POSITIVE 6 1 SENSITIVITY 62.5% SPECIFICITY 85.7% CYTOLOGY POSITIVE (C4,5) 3 5 C3 cytology and FISH Total tests = 21 4 C3 FISH negative C3 FISH positive 17 Future applications пЃ® пЃ® Emerging evidence that persistent positive FISH following BCG treatment is predictive of stage progression of bladder cancer Cystectomy may be offered to these patients following a course of BCG and positive FISH test Acknowledgments пЃ® The UroCyt vials used in this study were provided by Hologic.