Enhanced Endoscopy with IMAGE1 S CHROMA Improves Detection of Nonmuscle Invasive Bladder Cancer During Transurethral Resection

Abstract

Purpose:

We assessed the ability of enhanced cystoscopy with CHROMA image enhancement (Karl Storz SE & Co. KG, Tuttlingen, Germany) to improve cancer detection during transurethral resection of bladder tumors (TURBT) in patients with known or suspected nonmuscle invasive bladder cancer (NMIBC).

Patients and Methods:

A total of 49 patients (82% men) with a median age of 71 years underwent TURBT for known or suspected NMIBC. Bladders were assessed first with conventional white light imaging and then with CHROMA enhancement. We quantified the number of tumors seen with white light and any additional tumors seen with CHROMA and correlated with pathologic results.

Results:

A total of 165 tumors were viewed in 47 patients; of these, 25 were only seen using CHROMA. Use of CHROMA yielded additional tumors not appreciated on white light in 14 of 39 patients with underlying malignancy, for an additional detection rate (ADR) of 36% (95% confidence interval, 21%–53%). CHROMA improved detection of tumors in the setting of both low-grade (LG) and high-grade (HG) disease with equal efficacy (ADR 43% in LG and 44% in HG, p = 0.98, Barnard exact test).

Conclusions:

Cystoscopy with CHROMA contrast enhancement facilitates detection of bladder tumors that are not readily appreciated using conventional white light cystoscopy. The technology integrates seamlessly with current practice and can serve as an adjunct to other novel technologies.

Introduction

Urothelial Bladder Cancer carries a significant burden of morbidity and mortality rate, with >80,000 new diagnoses and nearly 18,000 deaths anticipated in the United States in 2020.¹ The mainstay of initial diagnosis and treatment in newly diagnosed cases is transurethral resection of bladder tumors (TURBT). In nonmuscle-invasive bladder cancer (NMIBC), which represents 70% to 80% of new cases, TURBT alone or in conjunction with intravesical treatments is frequently successful in eradicating disease. However, recurrence and progression risk is high, with both rates exceeding 50% after initial treatment in high-risk cases.² Incomplete resection of disease is common during TURBT, and in high-risk cases the rate of residual disease after initial TURBT can range from 25% to 75%.³ The outcome of TURBT is heavily dependent on technical factors, including the surgeon's ability to identify suspicious lesions and to perform a resection that is both wide enough to encompass the lesion and deep enough to sample detrusor muscle, while simultaneously avoiding bladder perforation. Several published series demonstrate that post-TURBT recurrence rates vary highly among individual surgeons and institutions and are dependent on surgeon experience.^4
–6

Given the importance of technical factors in determining oncologic outcomes after TURBT, there is a critical need for innovations that improve the technical quality of the procedure. Advances in this domain include narrow-band imaging (NBI) or photodynamic diagnosis (PDD) with blue-light fluorescent cystoscopy, both of which augment conventional white light imaging and have been shown to enhance detection of NMIBC and reduce recurrence rates.^7,8 However, both technologies require specialized imaging equipment that may not be available at all centers, and PDD requires preoperative intravesical instillation of hexaminolevulinate (HAL), which can increase costs and time spent in preoperative holding. In addition, the extirpative portion of TURBT is performed under white light, necessitating toggling between white and blue light for white light-negative lesions. One opportunity for further progress lies in computational image processing to enhance the standard white light image. In this study, we assessed the ability of one such image processing technology, CHROMA (Karl Storz SE & Co. KG, Tuttlingen, Germany) to increase detection rate of tumors during TURBT in suspected cases of NMIBC.

Patients and Methods

The study was structured as a single-institution, single-arm, open-label prospective trial combining conventional white light illumination and CHROMA-enhanced cystoscopy. Subjects were male and female patients between the ages of 18 and 90 years with a known or suspected (based on cystoscopic evaluation) diagnosis of NMIBC. Earlier intravesical treatment for NMIBC was permitted. The study protocol was reviewed and approved by the Institutional Review Board of the University of Texas—Southwestern Medical Center (IRB #112017-006) and the study was carried out in accordance with the Declaration of Helsinki (7th revision, Fortaleza, 2013). All patients provided written informed consent before enrollment. All study data and the decision to publish were in the hands of the investigators.

CHROMA image enhancement

Karl Storz SE & Co. developed a contrast-enhancing technology called CHROMA. The technology is described in detail elsewhere but, in brief, CHROMA incorporates data from a wide region surrounding each image pixel to increase contrast of structures such as blood vessels.⁹ This requires a much greater computational load than conventional edge enhancement but produces better contrast without the dark edges found in previous algorithms. Cystoscopic images produced with CHROMA are subjectively rated as higher quality than conventional white-light images, and CHROMA has been shown to compare favorably with NBI in detection of laryngeal tumors.^10,11 The CHROMA technology is an included feature of the IMAGE1 S™ camera system currently sold by Karl Storz. Use of the CHROMA feature requires the CONNECT™ and H3-LINK™ modules and H3P™ pendulum camera head.

Study design

Patients enrolled in the trial had a suspected diagnosis of new or recurrent NMIBC based on cystoscopic evaluation and were scheduled for TURBT in the operating room according to the usual standard of care. All resections were performed by a single expert urologic oncologist proficient in the procedure (author Y.L.). In the operating room, patients first underwent a comprehensive cystoscopy with conventional white light to identify suspect lesions, each of which was mapped on a standardized form. The bladder was then reevaluated using the CHROMA setting and each additional suspicious lesion was mapped. Each lesion was then biopsied or resected in its entirety using white light and hemostasis achieved. A second comprehensive cystoscopy was then performed with CHROMA enhancement, and any additional lesions were mapped and then biopsied or resected. Resections were performed using monopolar electrocautery with water or glycine as the irrigation fluid. All specimens were sent for routine pathologic analysis. Tumors detected with CHROMA alone were sent for analysis separately. Pathologic analysis was performed by fellowship-trained genitourinary pathologists at UTSW and grade was assigned according to the 2004 WHO classification system. For each patient, the number of tumors detected by white light and by CHROMA only were recorded, as well as the highest grade and stage of tumor detected by each modality. Subsequent management of patients was according to the usual standard of care.

Statistical analyses

The primary outcome of the study was the additional detection rate (ADR) of NMIBC, defined as the number of patients with additional tumors detected only by CHROMA divided by the total number of patients with tumors detected by any modality. Secondary endpoints included the rate of tumor upgrading/upstaging by CHROMA and the location of additional tumors detected by CHROMA. Information was collected in a secure database and descriptive statistics were obtained and reported.

Results

The trial enrolled a total of 50 patients undergoing TURBT, including two patients who underwent TURBT twice and were counted separately. One patient was subsequently found to be ineligible owing to a history of muscle invasive bladder cancer treated with chemoradiation, and was excluded from further analysis. Characteristics of the remaining 49 patients are given in Table 1. A majority of patients (31 of 49, 63%) had recurrent NMIBC; the remainder represented new cases of bladder cancer. Of patients with recurrent NMIBC, a majority (74%) had a history of high-grade (HG) disease. A large proportion (39%) of patients with recurrent disease had a prior diagnosis of carcinoma in situ, either with or without a concurrent papillary lesion. In addition, 18 of 31 (58%) and 7 of 31 (23%) patients had undergone prior treatment with Bacille Calmette–Guérin and/or intravesical chemotherapy, respectively. As detailed in Table 1, several patients had a history of upper tract urothelial carcinoma, prostate cancer, or prior pelvic radiation. Six of 49 patients were undergoing second-look TURBT, defined as a second resection shortly after an initial TURBT without intervening treatment.

Table 1.

Characteristics of Patients

Characteristic	Median (IQR) or n (%)
Sex
Male	40 (82)
Female	9 (18)
Age (years)	71 (65–76)
Race
White	46 (94)
Black	2 (4)
Hispanic	1 (2)
Prior urologic history
UTUC	6 (12)
PrCa	6 (12)
XRT	3 (6)
History of NMIBC
Yes (i.e., recurrent disease)	31 (63)
No	18 (37)
Prior pathology analysis
LG Ta	8 (26)
HG Ta	11 (36)
HG T1	12 (39)
CIS (any component)	12 (39)
Prior treatment
BCG	18 (58)
Intravesical chemotherapy	7 (23)
Second-look TURBT	6 (12)
Interval from initial TURBT (days)	53 (33–60)

Tumor grade is according to the 2004 WHO classification. CIS is reported as number of patients with any component of CIS, regardless of concurrent papillary lesions.

BCG = Bacille Calmette–Guérin; CIS = carcinoma in situ; HG = high-grade; IQR = interquartile range; LG = low-grade; NMIBC = nonmuscle invasive bladder cancer; PrCa = prostate cancer; TURBT = transurethral resection of bladder tumors; UTUC = upper tract urothelial carcinoma; XRT = pelvic radiation therapy.

Example images of tumors as seen under white light and CHROMA are provided in Figure 1. All tumors seen on white light were also visible with CHROMA. One patient had innumerable tumors throughout the bladder and was excluded from the analysis regarding the number of tumors detected. Another patient had no tumor seen on either modality—this patient had initially presented with gross hematuria and had several suspicious bladder lesions on office flexible cystoscopy; however, in the operating room, these were found to have resolved and were suspected to been an inflammatory process. He has had no further urologic problems in 1 year of follow-up. A total of 165 tumors were viewed in all patients, of which 140 were seen under white light and 25 were seen only with CHROMA (Table 2). The median number of tumors detected per patient with white light and CHROMA was 2 and 3, respectively (first and third quartiles, 1–4 and 2–4, respectively). Five patients had benign histology, one had prostate adenocarcinoma, and two patients had muscle-invasive (T2+) urothelial bladder cancer. The remainder had varying stages and grades of NMIBC as detailed in Table 2.

FIG. 1.

Examples of lesions as seen under conventional white light imaging and with CHROMA contrast enhancement. Representative examples of papillary and flat bladder lesions as seen under conventional white light imaging (top row) and with CHROMA image enhancement (bottom row).

Table 2.

Pathologic Characteristics of Tumors

	Median (IQR) or n (%)
Tumors observed (total, all patients)
White light	140
CHROMA	165
Tumors observed (per patient)
White light	2 (1–4)
CHROMA	3 (2–4)
Pathology analysis
No tumor seen	1 (2)
Benign	5 (10)
Prostate adenocarcinoma	1 (2)
LG Ta	15 (36)
HG Ta	13 (31)
HG T1	10 (24)
HG T2+	2 (5)
CIS (any component)	9 (21)

Grades are reported according to the 2004 WHO classification. CIS is reported as number of patients with any component of CIS (reported independently of any concurrent papillary lesion). The denominator for all percentages is 49, the total number of patients enrolled. There were no tumors seen with white light that were not also visible on CHROMA.

In four patients with NMIBC, tumors detected by white light and CHROMA were not sent separately for pathology analysis. Excluding these and patients with benign disease, a total of 14 of 39 patients had biopsy-proven malignant lesions detected with CHROMA, which were missed under white light, yielding an ADR of 36% (exact binomial 95% confidence interval, 21%–53%). CHROMA appeared to be especially efficacious in detecting lesions of the dome and anterior wall; at these sites nearly a quarter of lesions were detected by CHROMA only (Table 3). We inquired whether CHROMA would be more efficacious at detecting low-grade (LG) or HG lesions; however, we found no difference in the ability of CHROMA to detect additional lesions when cases were stratified by tumor grade. Additional tumors were detected in 43% of LG cases vs 44% of HG cases (Table 4; p = 0.98, Barnard exact test). In one case, a tumor (Ta LG) was detected with CHROMA when none were seen initially with white light. In the remainder of cases, additional lesions detected on CHROMA did not result in pathologic upstaging (data not shown).

Table 3.

Location of Additional Tumors Detected Only with CHROMA

	n/N (%)
Tumor location (CHROMA only/all tumors)
Trigone	0/6 (0)
Right ureteral orifice	2/11 (18)
Left ureteral orifice	2/19 (11)
Right lateral wall	2/28 (7)
Left lateral wall	2/16 (13)
Anterior wall	4/17 (24)
Posterior wall	6/35 (17)
Dome	5/19 (26)
Bladder neck	2/14 (14)
Urethra	0/0 (not calculable)

Percentages are reported as the number of tumors detected with CHROMA only divided by the total number of tumors detected at a given site.

Table 4.

Ability of CHROMA to Detect Additional Lesions Does Not Depend on Tumor Grade

	LG	HG	Total
(−) Additional on CHROMA	8 (57%)	15 (56%)	23
(+) Additional on CHROMA	6 (43%)	12 (44%)	18
Total	14	27	41

Contingency table plotting number of patients with additional lesions seen on CHROMA (rows) against 2004 WHO tumor grade (columns). p = 0.98 (Barnard exact test). Patient in whom entire bladder was affected was excluded. This analysis included patients in whom CHROMA-detected lesions were not sent separately from white light-detected lesions.

Median follow-up was 218 days (range, 8–374). Adverse events within 90 days of operation were noted in 10 of 49 (20%) patients, including one patient who experienced two complications. Minor complications included two patients with urinary tract infections, two with catheter-related problems, two with acute urinary retention, one with bothersome urinary frequency treated with mirabegron, and one with orthostatic dizziness that was attributed to tamsulosin. Major complications included one stroke, one postoperative ICU transfer for hypotension, and one readmission for acute appendicitis (deemed unrelated to the urologic procedure). There were no bladder perforations, technical errors, or other complications attributable to the use of CHROMA.

Discussion

TURBT is the cornerstone of management of NMIBC, as reflected by its central position in the guidelines of multiple major societies.^12
–14 Despite high variability in patient outcomes, the procedure remains recognizably the same as when it was first described nearly a century ago.¹⁵ This comparative absence of progress has led to calls for further innovation to improve the technical quality of the procedure and, ultimately, provide superior and more uniform patient outcomes.¹⁶ NBI and fluorescent PDD with HAL have both demonstrated the ability to reduce disease recurrence after TURBT^7,8,17
–19; however, both technologies require specialized equipment and frequent switching from enhanced to conventional white light modes during the course of a resection.

Enhancing the quality of the conventional white light image could represent a further refinement in TURBT that could be incorporated seamlessly alongside other technologies. The Karl Storz IMAGE1 S technology suite includes a digital enhancement mode called CHROMA, which is intended to enhance contrast and thus facilitate detection of bladder tumors. Images obtained with CHROMA were subjectively rated as superior to conventional images obtained with white light,¹⁰ and the technology demonstrated a comparable sensitivity and specificity to NBI in the detection of laryngeal tumors during laryngoscopy.¹¹

In the present prospective study, we assessed the ability of enhanced endoscopy with CHROMA to detect urothelial tumors that were missed by conventional white light. To our knowledge this represents the first systematic assessment of the utility of CHROMA in clinical use during TURBT. Use of CHROMA resulted in detection of additional tumors missed by conventional white light in 36% of patients, including one patient in whom no lesion was detected using white light. CHROMA appeared especially effective at detecting lesions in the bladder dome and anterior bladder wall, where roughly a quarter of lesions were seen with CHROMA only. CHROMA appeared to be equally efficacious in both HG and LG disease. Together, these findings suggest a promising role for CHROMA as an extension of conventional white light cystoscopy, possibly in conjunction with other enhancement modalities such as PDD. The use of CHROMA would require essentially no changes to usual practices in performing TURBT and could therefore be adopted seamlessly. In addition, resection can be performed in real time under CHROMA enhancement, whereas this is generally not done with NBI or PDD.

One major shortcoming of this study is its comparatively small number of patients. Another potential limitation is the sequential (white light, then CHROMA) design of the study, which could bias the findings in favor of CHROMA owing to improved tumor detection on a second assessment of the bladder. To avoid crediting CHROMA for false positives, we were careful to confine our calculation of ADR to cases in which tumors detected by CHROMA were pathologically demonstrated to be malignant. The single-surgeon design of our study has the benefit of eliminating interoperator variability, but also limits the generalizability of our findings. It is possible that the CHROMA system may be more or less beneficial in the hands of less experienced or nonspecialist surgeons. It is also important to emphasize that the CHROMA technology by itself cannot be expected to eliminate or compensate for the numerous other technical details that combine to result in a high-quality TURBT. In addition, the study was not designed to assess oncologic outcomes and it remains an open question as to whether the improved tumor detection rate seen in this study will result in an improvement in outcomes. Addressing these issues will require further prospective investigation in the setting of a randomized trial.

Conclusion

In this single-arm, single-institution prospective trial, the use of enhanced cystoscopy with CHROMA contrast enhancement resulted in increased detection of both LG and HG tumors in patients with suspected NMIBC.

Footnotes

Author Disclosure Statement

Karl Storz Endoscopy America, Inc. (El Segundo, CA) provided equipment (CHROMA-enabled cystoscopy cameras) for the study. Y.L. and S.D. serve as consultants for Photocure ASA, the manufacturer of HAL (Cysview^®/Hexvix^®). S.L.W. and J.M.H. report no conflicts.

Funding Information

Karl Storz provided equipment and editing services for the study. No external funding was received from Karl Storz or any other source.

Abbreviations Used

References

Siegel

, Miller

, Jemal

. Cancer statistics, 2020. CA. Cancer J Clin, 2020; 70:7–30.

Herr

. Natural history of superficial bladder tumors: 10- To 20-year follow-up of treated patients. World J Urol, 1997; 15:84–88.

Babjuk

. Second resection for non-muscle-invasive bladder carcinoma: Current role and future perspectives. Eur Urol, 2010; 58:191–192.

Brausi

, Collette

, Kurth

, et al. Variability in the recurrence rate at first follow-up cystoscopy after TUR in stage Ta T1 transitional cell carcinoma of the bladder: A combined analysis of seven EORTC studies. Eur Urol, 2002; 41:523–531.

Mariappan

, Finney

, Head

, et al. Good quality white-light transurethral resection of bladder tumours (GQ-WLTURBT) with experienced surgeons performing complete resections and obtaining detrusor muscle reduces early recurrence in new non-muscle-invasive bladder cancer. BJU Int, 2012; 109:1666–1673.

Rolevich

, Minich

, Nabebina

, et al. Surgeon has a major impact on long-term recurrence risk in patients with non-muscle invasive bladder cancer. Cent Eur J Urol, 2016; 69:170–177.

Burger

, Grossman

, Droller

, et al. Photodynamic diagnosis of non-muscle-invasive bladder cancer with hexaminolevulinate cystoscopy: A meta-analysis of detection and recurrence based on raw data. Eur Urol, 2013; 64:846–854.

Kang

, Cui

, Chen

, et al. Narrow band imaging-assisted transurethral resection reduces the recurrence risk of non-muscle invasive bladder cancer: A systematic review and meta-analysis. Oncotarget, 2017; 8:23880–23890.

Kamphius

, de Bruin

, Fallert

, et al. Storz professional image enhancement system: A new technique to improve endoscopic bladder imaging. J Cancer Sci Ther, 2016; 8:3.

10.

Kamphuis

, De Bruin

, Brandt

, et al. Comparing image perception of bladder tumors in four different Storz professional image enhancement system modalities using the SPIES App. J Endourol, 2016; 30:602–608.

11.

Staníková

, Walderová

, Jančatová

, et al. Comparison of narrow band imaging and the Storz Professional Image Enhancement System for detection of laryngeal and hypopharyngeal pathologies. Eur Arch Otorhinolaryngol, 2018; 275:1819–1825.

12.

Chang

, Boorjian

, Chou

, et al. Diagnosis and treatment of non-muscle invasive bladder cancer: AUA/SUO guideline. J Urol, 2016; 196:1021–1029.

13.

Babjuk

, Böhle

, Burger

, et al. EAU Guidelines on non–muscle-invasive urothelial carcinoma of the bladder: Update 2016. Eur Urol, 2017; 71:447–461.

14.

Woldu

, Bagrodia

, Lotan

. Guideline of guidelines: Non-muscle-invasive bladder cancer. BJU Int, 2017; 119:371–380.

15.

Gibson

. Treatment of bladder tumors with the McCarthy resectoscope. J Urol, 1935; 34:8–9.

16.

Mostafid

, Babjuk

, Bochner

, et al. Transurethral resection of bladder tumour: The neglected procedure in the technology race in bladder cancer. Eur Urol, 2020; 77:669–670.

17.

Hermann

, Mogensen

, Carlsson

, et al. Fluorescence-guided transurethral resection of bladder tumours reduces bladder tumour recurrence due to less residual tumour tissue in Ta/T1 patients: A randomized two-centre study. BJU Int, 2011; 108:297–304.

18.

Geavlete

, Multescu

, Georgescu

, et al. Treatment changes and long-term recurrence rates after hexaminolevulinate (HAL) fluorescence cystoscopy: Does it really make a difference in patients with non-muscle-invasive bladder cancer (NMIBC)?. BJU Int, 2012; 109:549–556.

19.

Gallagher

, Gray

, Anderson

, et al. ‘Real-life experience’: Recurrence rate at 3 years with Hexvix^® photodynamic diagnosis-assisted TURBT compared with good quality white light TURBT in new NMIBC—A prospective controlled study. World J Urol, 2017; 35:1871–1877.