Is restaging transurethral resection (TUR) necessary in patients with non-muscle invasive bladder cancer (NMIBC) and limited lamina propria invasion?

Abstract

Objectives

To evaluate the influence of lamina propria invasion type at initial transurethral resection (TUR) on re-staging pathology.

Materials and Methods

We reviewed prospectively-maintained records of all patients with a high-grade pT1 non-muscle invasive bladder cancer (NMIBC) who underwent both initial and restaging TUR within 6 weeks at our center between 2001 and 2016. The pathology of second TUR specimens was analyzed with regard to the characteristics of lamina propria invasion found at initial resection.

Results

We included 198 patients, with a median age of 70 years (interquartile range: 63–79). Muscle was present in the initial TUR specimen in 107 patients (54%). Pathology restaging was pT0 in 73 patients (37%), pTis in 44 (22%), pTa in 27 (14%), pT1 in 50 (25%) and pT2 in 4 (2%). Eighty-seven patients (44%) had tumors with minimal lamina propria invasion at initial TUR: 53 specimens (27%) had focal invasion (few malignant cells in the lamina propria); 15 specimens (7.6%) had superficial invasion (invasion of the lamina propria to the level of the muscularis mucosae (T1a)); and 19 specimens (10%) had multifocal superficial invasion (multiple areas of T1a). Of the patients with minimal lamina propria invasion, residual disease was found in 54 patients (62%). However, none of those patients had T2 disease.

Conclusions

A significant number of patients with T1 tumors have residual disease at restaging TUR as do patients with minimal lamina propria invasion. The extent of T1 invasion doesn't eliminate the need for repeat TUR.

Introduction

Urothelial carcinoma of the bladder affects more than 430,000 people worldwide each year, of which 75% have non-muscle invasive bladder cancer (NMIBC) confined to the mucosa (Ta or carcinoma in situ (CIS)) or lamina propria (T1) but not involving the muscularis propria [1]. Although these tumors are grouped under the heading of NMIBC, nearly all T1 tumors are high grade and have the potential to progress to muscle invasion and extravesical dissemination [2]. These high-risk NMIBC are often understaged by initial resection, with a risk of muscle-invasive disease detected by second resection from 4% to 25% [3]. It has been demonstrated that a second transurethral resection (TUR) plays a role in diagnosis, prognosis and treatment [4]. For these reasons, both the American Urological Association (AUA) and the European Association of Urology (EAU) guidelines recommend performing a second TUR for high-risk NMIBC [4,5].

Substaging of T1 tumors has been investigated to improve prognostic classification. In a recent meta-analysis, the depth of lamina propria invasion was shown to be the most prognostic factor in T1 NMIBC [6]. In this study, tumors with lamina propria invasion beyond the muscularis mucosae had an increased hazard for progression three-times greater than tumors with minimal involvement of the lamina propria. However, the influence on the re-staging pathology of lamina propria invasion type found at initial TUR has never been evaluated. In order to assess the benefit of a systematic second look, this study analyzed the results of a second TUR for T1 NMIBC with regard to the characteristics of lamina propria involvement at the time of the initial resection.

Materials and Methods

An institutional board-approved, retrospective review of the institutional database was performed to identify patients with clinical T1 (cT1) urothelial carcinoma of the bladder and no prior history of urothelial carcinoma, who underwent both an initial TUR and a re-staging TUR at our center. Diagnosis of cT1 disease was based on cystoscopy, bimanual examination, radiographic tests and pathologic evaluation. All tumors were staged according to the TNM 2009 system of the International Union Against Cancer and graded according to the World Health Organization / International Society of Urological Pathology 2004 grading system.

All pathology specimens were initially reviewed by a dedicated uro-pathologist. The characteristics of lamina propria invasion were assessed to distinguish minimal invasion from non-minimal invasion of the lamina propria. Minimal invasion of the lamina propria included: focal invasion, defined as few malignant cells in the lamina propria; superficial invasion, defined as invasion of the lamina propria to the level of the muscularis mucosae (T1a); and multifocal superficial invasion, defined as multiple areas of T1a. All other types of invasions into or beyond the muscularis mucosae were categorized as non-minimal.

Intravesical therapy was not administered after the first TUR. The second TUR was performed within 6 weeks after the first one. During the second TUR, all visible tumors and previous resection scars were resected. Residual tumor was defined as restaging higher than pT0. The pathology of the second TUR was analyzed with regard to the characteristics of the initial resection.

We used Fisher’s exact and chi-squared tests to examine the relationship between depth of tumor invasion at initial TUR and presence of residual disease, with p < 0.05 considered to be significant. We also used chi-squared tests to study the association between residual disease and presence of muscle tissue in the initial TURB. All analyses were performed using Stata 13 (StataCorp, College Station, TX).

Results

We identified 198 patients with initial cT1 disease who underwent a re-staging TUR within 6 weeks of their initial surgery from 2001 to 2016 (Table 1). The patients’ median age was 70 years (interquartile range: 63 – 79 years) and 74% of the sample was male. Initial TUR was macroscopically complete in all cases. All pT1 tumors were high grade and muscle was present in the initial TUR specimen in 107 patients (54%). Eighty-seven patients (44%) had tumors with minimal lamina propria invasion at initial TUR: 53 specimens (27%) contained focal invasion, 15 specimens (7.6%) had superficial invasion and 19 specimens (10%) had multifocal superficial invasion.

Table 1.

Patient and TUR characteristics

Variables	N=198
Median age, years (interquartile range)	70 (63–79)
Male	146 (74%)
Muscle present in TUR 1	107 (54%)
TUR 1 Lamina Propria Invasion:
Focal Invasion	53 (27%)
Superficial Invasion	15 (7%)
Multifocal Superficial Invasion	19 (10%)
Non-minimal invasion	111 (56%)
Muscle present in TUR 2	168 (85%)
TUR 2 Pathology:
pT0	73 (37%)
pTis	44 (22%)
pTa	27 (14%)
pT1	50 (25%)
pT2	4 (2%)

Pathology restaging was pT0 in 73 patients (37%), pTis in 44 (22%), pTa in 27 (14%), pT1 in 50 (25%) and pT2 in 4 (2%). Muscle was identified in 168 specimens (85%). Overall, 125 patients (63%) had residual disease. Of those with minimal lamina propria invasion, residual disease was found in 54 patients (62%); however, none of those patients had T2 disease (Table 2). There was no significant difference in the risk of residual disease according to lamina propria invasion type at initial TUR (Table 3). The presence of muscle in the initial staging specimens was significantly associated with fewer incidences of residual disease (Table 3). However, 21 patients (52%) with minimal lamina propria invasion and presence of muscle at initial TUR still had residual disease at restaging, including 6 patients (15%) whose tumors were pT1 at restaging.

Table 2.

Pathology of repeat TUR specimens, stratified by invasion type at initial TUR

	Invasion type at initial TUR
Pathology at repeat TUR	Non- minimal invasion N=111 (56%)	Minimal invasion
		Focal Invasion N=53 (27%)	Superficial Invasion N=15 (7%)	Multifocal Superficial Invasion N=19 (10%)
pT0	40 (36%)	21 (40%)	6 (40%)	6 (31%)
pTis	22 (20%)	15 (28%)	4 (27%)	3 (16%)
pTa	15 (13%)	4 (8%)	1 (6%)	7 (37%)
pT1	30 (27%)	13 (24%)	4 (27%)	3 (16%)
pT2	4 (4%)	0 (0%)	0 (0%)	0 (0%)

Table 3.

Presence of residual disease in the restaging specimen according to lamina propria invasion type and according to the presence of muscle at initial TUR

	Residual disease	No residual disease	p
Minimal lamina propria invasion	54 (62%)	33 (38%)	0.74
Non-minimal lamina propria invasion	71 (64%)	40 (36%)
Presence of muscle	62 (57%)	47 (43%)	0.04
No muscle	63 (71%)	26 (29%)

Discussion

TUR is a crucial procedure in the diagnosis and treatment of NMIBC. A complete and correctly performed TUR is essential to achieve good prognosis: a significant risk for residual tumor after initial TUR of T1 lesions has been demonstrated [7]. In our study, we found that only 37% of the patients were pT0 at restaging and 27% had pT1 or above residual disease, including 2% who had muscle-invasive disease. These patients with unexpected residual disease would have risked progression while on BCG. Many studies have reported that around 50% of patients newly diagnosed with cT1 tumors presented with residual tumor at restaging TUR within 6 weeks after initial resection (Table 4) [8–20]. According to a meta-analysis conducted by Vianello et al, the residual rate at second TUR is 47% (95% CI = 0.41–0.53) [21]. Furthermore, the likelihood that muscle-invasive disease is detected by repeat resection of an initially T1 tumor ranges from 4% to 25%; if no muscle is present in the initial resection, the risk increases to 45% [3]. This risk increased to 50% in some radical cystectomy series, although these studies only enrolled selected patients [22]. Based on these findings, various guidelines strongly recommend a second TUR for patients newly diagnosed with cT1 tumors [4,5]. A restaging TUR has 4 objectives: diagnostic, therapeutic, prognostic and predictive. Indeed, it has been demonstrated that a second TUR can increase recurrence-free survival, provide prognostic information, and predict outcomes after BCG treatment [23,24].

Table 4.

Pathological results of restaging TUR for T1 NMIBC at initial resection

Study	N	Residual tumor	Muscle invasion	Delay after initial TUR (weeks)
Klän et al. 19918	46	43%	2%	1–2
Herr et al. 19999	58	78%	28%	4–6
Brauers et al. 200110	42	64%	5%	4–6
Schips et al. 200211	76	33%	8%	4–6
Schwaibold et al. 200612	136	52%	10%	4–6
Divrik et al. 201013	105	33%	8%	2–6
Ali et al. 201014	91	67%	26%	2–6
Yucel et al. 201015	33	35%	11%	3–6
Katumalla et al.201116	50	36%	4%	4–6
Takaoka et al. 201317	73	51%	4%	< 12
Lazica et al. 201318	167	58%	7%	4–6
Shim et al. 201519	29	76%	10%	3–8
Gontero et al. 201620	935	71%	Excluded	4–6
Present study	198	63%	2%	< 6

In the EORTC risk tables, the T1 high-grade tumors do poorly, with 1- and 5-year disease progression rates of 11.4% and 19.8%, respectively [2]. However, cT1 tumors are heterogeneous with respect to recurrence and prognosis. Some tumors have the potential to progress to muscle-invasive or metastatic disease and these patients should be identified as candidates for aggressive treatment with early cystectomy. Additional prognostic factors have been described in selected patient populations. Female sex, age, tumor size, concomitant CIS and CIS in the prostatic urethra are important prognostic factors in T1 patients treated with BCG [25]. In a recently published international, multicentric, retrospective study, re-TUR in the presence of muscle in the primary specimen didn’t improve the outcomes, and re-TUR in the absence of muscle had a borderline significant positive impact on time to recurrence, time to progression, overall survival and cancer specific survival [20]. However, in a randomized study, patients with T1 disease at first TUR had a two-fold increased risk of recurrence and a four-fold increased risk of progression when re-TUR was not done [23].

Aiming to risk stratify T1 bladder cancers, several substaging systems have been proposed. In particular, the extent of the lamina propria invasion has been investigated. Several reports have explored the utility of examining invasion of the muscularis mucosae (MM) layer to substage pT1 tumors. The T1a/T1b/T1c system holds prognostic value for progression and disease-specific survival. T1a indicates invasion up to the MM; T1b, invasion into the MM; and T1c, invasion beyond the MM. The meta-analysis performed by Martin-Doyle et al based on 1,431 patients identified depth of invasion into lamina propria (T1b/c) as the risk factor with the highest impact: they found a progression hazard ratio [HR] = 3.34 (p < 0.001) and a cancer-specific survival HR = 2.02 (p < 0.001) [6]. In a prospective study, pT1 substaging was used to determine the indication of re-TUR after BCG [26] and in multivariate analysis, substaging, CIS and tumor size were significant predictors of progression. Recent evidence suggests that histopathologic classification into T1a and T1b can be translated into a molecular signature, reflecting progressive dysregulation and resulting in more invasive stages [27]. However, the main problem with this system is the lack of consensus regarding identification of the MM at the point of invasion of the tumor, causing a variable substaging rate (58–100%). Reasons for the variation may include the difference in quality of TUR, damage to the lamina propria by previous TUR, and intravesical treatment.

However, even when minimal invasion of the lamina propria is identified, residual tumor is frequently discovered at restaging TUR. In our series, residual disease was found in 60% of patients who had focal invasion at initial TUR, which is similar to findings in tumors with superficial invasion (60%), multifocal superficial invasion (69%) or non-minimal invasion (64%). Different patterns of T1 tumors may affect margins of resection and consequently residual tumor left behind after initial TUR [3]. Some tumors have broad-front invasion whereas other present tentacular invasion, requiring removal of not only the visible epithelial portion of tumors, but also the invisible microscopic mucosal or invasive tumor, including deep muscle [3]. Therefore, because the type of lamina propria invasion remains unpredictable during TUR, we suggest that all T1 tumors should undergo restaging TUR, even if the initial specimen presents with minimal invasion.

In order to optimize TUR, photodynamic diagnosis (PDD) including fluorescence cystoscopy and narrow-band imaging has demonstrated a better sensitivity than conventional procedures for the detection of tumors [28]. A meta-analysis by Burger et al reported an increase in detection of tumor lesions and an absolute reduction of 11% in recurrence rates within 12 months [29]. However, the benefit on progression rate and survival remains to be demonstrated and no data is available regarding the rate of residual tumor at restaging within 6 weeks.

A potential limitation of this study is its retrospective design. We focused our study on limited invasion of the lamina propria, including focal invasion, superficial invasion, and multifocal superficial invasion. We didn’t specifically analyze other prognostic factors such as multifocality, tumor size, CIS or smoking status, nor the effects on long-term recurrence and progression rate. Although all the specimens were reviewed by a dedicated uro-pathologist, the description of the type of lamina propria invasion didn’t use the previously published substaging systems. Furthermore, the presence of muscle reported at initial TUR was surprisingly low (54%) and may have multifactorial explanations, including perioperative visibility, surgical technique and pathological analysis. However, the low rate of pT2 at restaging (2%) with good muscle representation (85%) can be considered a marker of the quality of the initial resection. Finally, the aim of this study was to answer to a simple, clinically-relevant question: does limited invasion of the lamina propria reduce the need for restaging TUR? The answer is no, and a second resection should be performed every time an invasive tumor is diagnosed, due to the high risk of residual disease.

Conclusions

A significant number of patients with T1 tumors have residual disease at restaging TUR. This is not any different among patients with minimal lamina propria invasion. We suggest that the extent of T1 invasion doesn't eliminate the need for repeat TUR, for diagnostic, therapeutic, prognostic and predictive purposes. Subsequent treatments depend on the quality of the information provided by re-staging TUR. Molecular profiling of NMIBC for tailored treatment would then be the next step forward.

Highlights.

• Minimal lamina propria invasion doesn’t reduce the risk of residual disease.

• Residual disease can be found even if muscle was present at initial specimen.

• The extent of T1 invasion doesn't eliminate the need for repeat TUR.