SEARCH

SEARCH BY CITATION

Keywords:

  • penile carcinoma;
  • histological grading;
  • histological staging;
  • prognostic factors;
  • interobserver variation

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

OBJECTIVE

To examine interobserver variations in assessing grade and stage of penile squamous cell carcinoma (SCC).

PATIENTS AND METHODS

We retrospectively reviewed the pathological features and clinical outcome in 75 patients with SCC of the penis, who were treated in participating urological centres between 1996 and 2005; the assessments of the local pathologists and the review pathologists were compared.

RESULTS

There was conformity in tumour grade in 67% and the assessment of tumour stage conformed in 84%; the combination assessment of both grade and stage conformed in 56%.

CONCLUSION

Accurate histological subtyping by the surgical pathologist demands standardized guidelines, in particular for histological grading, which is crucial for clinical treatment, but shows significant interobserver variation.


Abbreviations
LNM

lymph node metastases

SCC

squamous cell carcinoma

EAU

European Association of Urology

Cis

carcinoma in situ.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The presence, extent and adequate therapy of inguinal lymph node metastases (LNMs) in patients with penile carcinoma are all crucial determinants of the survival rate [1–4]. Patients with penile carcinoma benefit from the early removal of occult LNM. However, lymph node dissection as a standard procedure in all patients leads to significantly many unnecessary surgical interventions, accompanied by severe morbidity rates [5,6].

To reduce the number of unnecessary lymph node dissections, several attempts have been made to predict nodal involvement in squamous cell carcinoma (SCC) of the penis. Histopathological staging and grading of penile SCC is important in characterizing the malignant and metastatic potential of this rare tumour. However, comparison of different studies shows highly diverging results for the risk of lymph node involvement in identical tumour stages or grade.

Slaton et al.[7] observed no lymph node involvement in 15 T1 carcinomas, whereas Theodorescu et al.[8] found a 58% risk of metastatic disease in 24 patients with T1 carcinoma, irrespective of the tumour grade. Ficarra et al.[9] recently established a nomogram to predict lymph node involvement, in a multicentre approach. According to this nomogram, intermediately differentiated tumours are at greater risk of LNM than poorly differentiated tumours. These results contradict other studies and were recently critically discussed by Leijte and Horenblas [10]. The combination of tumour stage and grade form the basis of risk classification according to the only available guidelines for penile SCC, published by t he European Association of Urology (EAU) [11].

To the best of our knowledge, there has been no central histopathological review of penile SCC. As penile SCC is a rare malignant disease in developed countries, accounting for only 0.2–0.6% of all male malignancies [1], it is not usually well known by pathologists and clinicians. This lack of experience compromises the assessment of the histopathological features, which in turn determine the patient’s treatment and prognosis.

Thus the aim of the present study was to examine interobserver variations in assessing the grade and stage of penile SCC, and to show how these affect the choice of therapy and prognosis.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The records of all patients with penile SCC from the participating urological centres in 1996–2005 were reviewed retrospectively, by two pathologists re-evaluating the respective tumour specimens. Patients were considered for this study if: (i) ethical approval requirements were fulfilled; (ii) tumour-free surgical margins were achieved by surgical intervention; (iii) all tumour specimens were available for re-assessment; (iv) the lymph node status was available (based on surgical staging or uneventful follow-up of >2 years); (v) there was no local tumour recurrence; and (vi) the TNM classification of the initial histopathology was done according to the current TNM guidelines [12].

The tumours were re-assessed by an experienced pathologist (G.K., 38 years of experience and Head of Department) and a well-trained third-year resident in general pathology (I.A.) who were both unaware of any clinical data. The histological re-assessment by two pathologists (according to TNM and WHO classifications [12,13]) of tumour grading and staging was based on light microscopy of the original slides of all tumour specimens. If there was disagreement between the initial evaluation of the two review pathologists (six tumours for stage and 13 for grade), a consensus was achieved at a multi-headed microscope. If there were damaged or missing slides, 4-µm thick sections of formalin-fixed, paraffin-embedded tissue from the original tumour specimens were stained with haematoxylin and eosin.

According to the above classifications, carcinoma in situ (Cis) is an early form of the carcinoma with any level of dysplasia, defined by the absence of invasion of the surrounding tissue. T1 penile tumours invade the basement membrane and there is a growth of islands, cords, or single tumour cells in the subepithelial stroma. T2 tumours invade the corpus spongiosum or the corpus cavernosum, while T3 tumours enter either the urethra or prostate. Tumours invading other adjacent structures are classified as pT4 tumours.

For grading, well-differentiated tumours show cytological and histological features similar to those of the normal penile squamous epithelium. In well-differentiated penile SCCs, there is a high proportion of large, differentiated, keratinocyte-like squamous cells and a low proportion of small basal-type cells, located in the periphery of the cancer cell nests. The presence of keratinization is interpreted as a sign of differentiation; mitoses are scanty. The architectural disturbance is limited to the lower third of the epithelium.

Moderately differentiated carcinomas are ranked between the well and poorly differentiated types and have no generally accepted criteria. The architectural disturbance extends into the middle third of the epithelium, accompanied by an upgraded cytological atypia. They have more nuclear pleomorphism and more mitosis, including abnormal mitoses. There is usually less keratinization.

Poorly differentiated tumours predominantly consist of basal-type cells, which usually have a high mitotic rate, including abnormal mitoses. The architectural disturbance is greater than two-thirds of the epithelium with associated cytological atypia or architectural disturbance in the middle third of the epithelium [13,14].

Risk groups for LNMs of penile carcinomas were established as suggested by the EAU [11]; pTis, pTaG1-2 or pT1G1 tumours form the low-risk group, pT1G2 the intermediate-risk group, and pT ≥2 or G3 are considered as high-risk tumours.

The histopathological assessment of the initial pathologist was termed ‘initial’ pathology, and the consensus assessment of the two review pathologists ‘review’ pathology. Differences in staging, grading and risk group between initial and review pathologists were tested for significance using Bowker’s symmetry test, with statistical significance defined at P < 0.05.

To assess the extent of agreement between the different observers, we used Cohen’s κ; the degree of agreement beyond chance was classified as: 0 as ‘none’, 0–0.2 as ‘slight’, 0.2–0.4 as ‘fair’, 0.4–0.6 as ‘moderate’, 0.6–0.8 as ‘substantial’ and 0.8–1 as ‘almost perfect’. Furthermore, the incidence of LNMs and overall survival was investigated in relation to staging, grading and risk groups; overall survival was depicted by Kaplan-Meier plots.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

In all, 75 patients were enrolled in the study (mean age 64 years, range 34–91; mean follow-up 35 months, range 0–142). The tumour grading as identified by initial and review pathologists is shown in Table 1. The conformity of grading assessment was 67%. The Cohen κ was 0.458, suggesting moderate agreement. There was no significant difference between initial and review pathology (P = 0.801). Compared to the review pathology results, the tumours were over-graded in 14% and under-graded in 19% by the initial pathologists. The incidence of lymphatic spread in different tumour grades, as found by initial and review pathology, is shown in Fig. 1.

Table 1.  Tumour grading and staging according to initial and review pathologists
TumourReview pathology
Initial pathology      
GradingG1G2G3  Total
 G111 4  15
 G2 42710  41
 G3 611  17
 Total153721  73
StagingCisT1T2T3T4 
 Cis1 1 2
 T131 839
 T2 125228
 T35 5
 T41 1
 Total 133337175
image

Figure 1. Incidence of LNM according to the grading of initial (green columns) and review (red columns) pathologists.

Download figure to PowerPoint

Both histopathological assessments show an increasing rate of metastatic disease for well, moderately and poorly differentiated tumours. In particular, the rate of LNMs in G1 tumours identified by initial pathologists was 20%, vs 7% by review pathologists. In moderately differentiated tumours the rate of lymphatic spread was 44%, vs 49%, and 65% vs 59% in poorly differentiated tumours (initial vs review pathology results).

The overall survival in relation to tumour grading is shown in Fig. 2. The initial and review pathology survival data differed for well-differentiated tumours, showing a clear association between the nodal status and survival. Two of 75 tumours were staged as ‘Cis’ by the initial pathologists and therefore not considered for comparison of grading.

image

Figure 2. Overall survival with grading (G1 green; G2 yellow; G3 red) as assessed by initial (solid line) and review (dotted line) pathologists.

Download figure to PowerPoint

Table 1 also shows the tumour stage assessment by initial and review pathology; Cohen’s κ for tumour stage was 0.733 (substantial agreement). The Bowker symmetry test for staging showed no significance (P = 0.586). The conformity of staging was 84%; according to review pathology 15% had been under- and 1% over-staged by initial pathologists. The most outstanding divergence of histopathological assessment was for T1 carcinomas; of 39 tumours staged as T1 by initial pathologists, eight were staged as T2 by review pathology. This change from T1 to T2 led to a decreased metastatic risk in T1 carcinomas of 27% in review staging, vs 36% in initial staging. The incidence of LNMs in T2 carcinomas was then 58% in review staging vs 57.1% in initial staging (Fig. 3). Figure 4 shows the overall survival for each tumour stage, as assessed by initial and review pathologists for T1 and T2 tumours. Survival for Cis, T3 and T4 tumours is not shown as there were too few patients in initial (eight) and review (nine) pathology.

image

Figure 3. LNMs in different stages. Initial (green columns) vs review pathology (red columns). *There were no metastases in Cis.

Download figure to PowerPoint

image

Figure 4. Comparison of survival in T1 (black) and T2 (red) carcinomas for initial (solid line) and review (dotted line) pathology. Survival for Cis, T3 and T4 tumours is not shown as there were too few patients for initial (eight) and review (nine) pathology in the study.

Download figure to PowerPoint

An overview of initial and review staging and grading of the 75 tumours is shown in Table 2; the combined conformity of grading and staging was 56%. These data were not tested for agreement (Cohen’s κ) or symmetry (Bowker’s symmetry) as symmetry was not present.

Table 2.  Staging and grading according to initial and review pathologists
Staging and gradingReview pathologyTotal
CisT1G1T1G2T1G3T2G1T2G2T2G3T3G1T3G2T3G3T4G2
Initial pathology            
 Cis11 2
 T1G1 8 311
 T1G2 3122 3222
 T1G3 12 12 6
 T2G111 2
 T2G2 1110517
 T2G3 351 9
 T3G111 2
 T3G211 2
 T3G31 1
 T4G31 1
Total11117521714223175

The risk group classification of the 75 tumours according to the EAU guidelines [1] is shown in Table 3, both for initial and review pathology. Cohen’s κ was 0.631 (substantial agreement), and Bowker’s symmetry test showed no significant difference between initial and review assessment (P = 0.286). The risk group assessment of initial and review pathology was congruent in 79%. According to review pathology, the risk groups had been underestimated in 15% and overestimated in 7% by initial pathology. The most outstanding differences between initial and review assessment were for the low- and intermediate-risk groups. According to review pathology, four of the 13 low-risk tumours and seven of the 22 intermediate-risk carcinomas (32%) had been underestimated by initial pathologists, while three of the 22 intermediate-risk tumours (14%) and two of 40 high-risk tumours (5%) had been overestimated by initial pathologists. For nodal involvement (Fig. 5), re-assessment led to a more distinctive rate of lymphatic spread, especially in the low-risk group, with 15% in initial and none in the review assessment. As already shown for grading and staging, nodal involvement had a direct effect on survival (Fig. 6).

Table 3.  EAU risk group assessment [11] according to initial and review pathologists
EAU risk groupReview pathologyTotal
LowIntermediateHigh
Initial pathology    
 Low 9 3 113
 Intermediate 312 722
 High 0 23840
Total12174675
image

Figure 5. The incidence of LNMs in the different risk groups according to the EAU guidelines [11] for initial (green columns) and review (red columns) assessment. *No metastases within the low risk group were found by review pathologist.

Download figure to PowerPoint

image

Figure 6. Overall survival in the EAU risk-groups (low-risk, green; intermediate-risk, yellow; high-risk, red) for initial (solid line) and review (dotted line) assessment.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The presence, extent and adequate therapy of inguinal LNMs in patients with penile SCC are all crucial determinants of the survival rate [1,4–6]. The presents study is the first to investigate the conformity of tumour grading, staging and the combination of both in penile SCC.

The incidence of lymphatic spread and the overall survival within the different T stages, tumour grades and risk groups served as a quality control for the initial and review histopathological assessment.

There are few studies describing the metastatic risk for the T stages according to the current TNM classification [15–17]. The incidence of nodal disease in T1 carcinomas was slightly higher in the present study, both as assessed by initial (36%) and review pathologists (27%), than in the findings of Maiche et al.[16] in 122 patients (25%). The occurrence of nodal spread in T3 carcinomas was 20% in initial vs 43% in the review assessment, which is even lower than the risk seen in T2 carcinomas (57% in initial and review assessment). Leitje et al.[15] found nodal spread in nine of 13 patients with T3 tumours, while nine of 12 patients developed LNM in the study of Maiche et al.[16]. The few patients staged T3 (five initial, seven review) and urethral infiltration due to location close to the urethral orifice in four patients might account for few with nodal disease.

The comparison between review and initial assessment showed substantial agreement in the staging of penile carcinomas (Cohen’s κ 0.733). This shows that the anatomical borders dividing the subepithelial connective tissue from the corpus spongiosum or cavernosum are identified in most cases, almost independent of the observer. Although there was no statistical difference in staging (P = 0.586) with a conformity rate of 84%, there was a clear tendency of initial pathologists to under-stage (15%) rather than to over-stage penile carcinomas (1%), which applies to T1 and T2 tumours especially. Hence, the presumed increasing incidence of LNM and shorter overall survival for T2 than T1 carcinomas has been identified more accurately by review assessment.

The incidence of LNM in relation to tumour grade has been shown to be 14–47% for well-, 37–66% for moderately, and 29–82% for poorly differentiated tumours [2,3,15,17–19]. The initial and review assessment were consistent for lymphatic spread in the different tumour grades, except for well-differentiated tumours, where a slightly lower rate (7%) of LNMs was found on re-assessment.

Conformity (67%) between initial and review grading was ‘moderate’ (Cohen’s κ 0.458). Initial pathologists tended to under-grade (19%) rather than to over-grade (14%), but the differences were not significant (P = 0.801).

The clinical validity of any histological grading system depends on the correlation between the biological behaviour of the lesions and the pathological assessment of tumour differentiation. There is no generally accepted histological grading system for penile SCC. Even if pathologists analyse the histology according to the above-mentioned criteria, there is no universally accepted score system required for an adequate diagnostic assessment. A first general grading system for SCC was established by Broders in 1921 [20] and modified by several authors to include various architectural and cytological changes [21–24].

However, as there are no generally accepted criteria to score the relative contribution of the different grading variables, the grading of SCC is subject to substantial interobserver variation [20–25]. The heterogeneous histopathological results due to the unfixed guidelines for histological grading compromise the selection of adequate therapeutic procedures.

Because of a similar dilemma with breast cancer, Elston and Ellis [26] initiated a semiquantitative evaluation of the morphological features of invasive adenocarcinomas of the breast by setting up a numerical scoring system to make the criteria more objective. The assessment of tumour grade in breast cancer is based on a summary score of the tubule formation, nuclear pleomorphism and mitotic counts (1–3 points for each variable). Equivalent morphological features could be defined for penile cancer and SCCs in general.

To minimize unnecessary lymph node dissections, the EAU proposed guidelines for assessing which patient groups are at risk of LNM by using a combination of stage and grade [11]. There are four main studies reporting the incidence of LNMs for the three risk groups. The rates for LNM were 0–6% for low-risk, 0–44% for intermediate-risk and 31–63% for high-risk carcinomas [2,3,15,17,18]. Except for the initial pathology finding of 20% nodal involvement in low-risk tumours, the results of the present study, both initial and review assessment, are consistent with the above-mentioned reports. The conformity between initial and review assessment for the EAU risk groups was 78% (Cohen’s κ= 0.631), with a tendency towards under-grading and/or -staging (15%) rather than over-grading and/or -staging (7%) by initial pathology. However, the review pathology showed a more distinctive graduation of nodal involvement and overall survival within the three risk groups.

The present study confirms that staging and grading of penile SCC is associated with substantial variation depending on the pathologist. Although the agreement between initial and review pathologists was ‘substantial’ for staging, and ‘moderate’ for grading, there was conformity for both variables in only 56%.

The statistic validity of study results is limited due to the rarity of penile SCC and considerable interobserver variation, which probably accounts for divergent and even contradictory findings, especially for the risk of LNMs within the different risk groups. In 2006 a nomogram predicting the probability of lymph node involvement suggested a greater effect of moderately differentiated and superficially spreading tumours, compared to poorly differentiated and vertically growing tumours [9]. As discussed by Leitje et al.[10], this contradicts recent findings. As there were relatively many patients (175), it must be assumed that these contradictory results arose because the histopathology was assessed in 11 centres by different pathologists with varying degrees of specialization. Furthermore, the present study shows that the incidence of LNMs and the survival within different tumour stages, tumour grades and risk groups was clearer and graded in review pathology assessment.

The diagnostic approach for intermediate-risk tumours according to the EAU guidelines [11] requires special comment. In the initial and review assessment the metastatic risk was respectively 45% and 41% for the intermediate-risk group, which is almost as high as in the high-risk group (50% and 54%, respectively). Recently published abstracts and reports discussing the incidence of LNMs in smaller cohorts of intermediate-risk patients showed overall metastatic risks of 0–66%[17,27,28]. This implies that the assessment of pT1G2 tumours as ‘intermediate-risk’ carcinoma might be an inaccurate approach.

The present study shows that objective and reliable guidelines for histological subtyping are needed to predict lymphatic involvement and to select adequate therapeutic procedures. Although the present study included relatively many patients (75) with this rare disease, our results require further confirmation through a re-evaluation of even larger patient groups, ideally by several pathologists in a multi-institutional approach.

Tumour stage and grade play a key role in predicting the risk of LNMs in penile SCC. The predictive value of tumour staging, grade and risk-group assessment is still limited due the rarity of the disease and interobserver variation. Controversial findings of different groups for the risk of metastatic spread within the same tumour stages and/or grades implies that surgical pathology needs to be improved and standardized to help the urologist to select adequate and safe treatment strategies.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

We thank Mrs Ulrike Schulz for her statistical advice, Professor Holger Kalthoff, Division of Molecular Oncology, University Hospital Schleswig Holstein, Campus Kiel for his scientific advice and Mrs Almut Kalz for proof-reading this manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES
  • 1
    Culkin JC, Beer TM. Advanced penile carcinoma. J Urol 2003; 170: 35965
  • 2
    Solsona E, Iborra I, Ricos JV et al. Corpus cavernosum invasion and tumour grade in the prediction of lymph node condition in penile carcinoma. Eur Urol 1992; 22: 1158
  • 3
    Solsona E, Iborra I, Rubio J, Casanova JL, Ricos JV, Calabuig C. Prospective validation of the association of local tumour stage and grade as predictive factor for occult lymph node micrometastasis in patients with penile carcinoma and clinically negative inguinal lymph nodes. J Urol 2001; 165: 15069
  • 4
    Sanchez-Ortiz RF, Curtis A, Pettaway CA. The role of lymphadenectomy in penile cancer. Urol Oncol 2004; 22: 23644
  • 5
    McDougal WS. Carcinoma of the penis: improved survival by early regional lymphadenectomy based on the histological grade and of invasion of the primary lesion. J Urol 1995; 154: 13646
  • 6
    Kroon BK, Horenblas S, Lont AP, Tanis PJ, Gallee MPW, Nieweg OE. Patients with penile carcinoma benefit from immediate resection of clinically occult lymph node metastases. J Urol 2005; 173: 8169
  • 7
    Slaton JW, Morgenstern N, Levy DA et al. Tumour stage, vascular invasion and percentage of poorly differentiated cancer: independent prognosticators for inguinal lymph node metastasis in penile squamous cancer. J Urol 2001; 165: 113842
  • 8
    Theodorescu D, Russo P, Zhang ZF, Morash C, Fair WR. Outcomes of initial surveillance of invasive squamous cell carcinoma of the penis and negative nodes. J Urol 1996; 155: 162631
  • 9
    Ficarra V, Zattoni F, Artibani W et al. Nomogram predictive of pathological inguinal lymph node involvement in patients with squamous cell carcinoma of the penis. J Urol 2006; 175: 17005.
  • 10
    Leijte JAP, Horenblas S. Rebuttal from Authors re. Eric Wespes. The management of regional lymph nodes in patients with penile carcinoma and reliability of sentinel node biopsy. Eur Urol 2007; 52: 15–6 and re: Vincenzo Ficarra and Antonio Galfano. Should the dynamic sentinel node biopsy (DSNB) be considered the gold standard in the evaluation of lymph node status in patients with penile carcinoma? Eur Urol 2007 52: 17–9. Eur Urol 2007; 52: 201
  • 11
    Solsona E, Algaba F, Horenblas S, Pizzocaro G, Windahl T. Guidelines on Penile Cancer. European Association of Urology. Update March 2004. Available at: http://www.uroweb.org/files/uploaded_files/guidelines/22891_Penile_Cancer.pdf. Accessed March 2004
  • 12
    Sobin LH, Wittekind Ch. TNM Classification of Malignant Tumours, 6th edn. New York: Wiley-Liss, 2002
  • 13
    Cubilla AL, Dillner J, Schellhammer PF et al. Tumours of the penis. In EbleJ, SauterG, EpsteinJI, SesterhennIA eds, World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs. Lyon: International Agency for Research on Cancer (IARC) Press, 2004: 27990
  • 14
    Gale N, Kambic V, Michaels L et al. The Ljubljana classification: a practical strategy for the diagnosis of laryngeal precancerous lesions. Adv Anat Pathol 2000; 7: 24051
  • 15
    Leijte JAP, Kroon BK, Valdes Olmos RA, Nieweg OE, Horenblas S. Reliability and safety of current dynamic sentinel node biopsy for penile carcinoma. Eur Urol 2007; 52: 1707.
  • 16
    Maiche AG, Pyrhonen S, Karkinen M. Histological grading of squamous cell carcinoma of the penis: a new scoring system. Br J Urol 1991; 67: 5226
  • 17
    Hegarty PK, Kayes O, Freeman A, Christopher N, Ralph DJ, Minhas S. A prospective study of 100 cases of penile cancer managed according to European Association of Urology guidelines. BJU Int 2006; 98: 52631
  • 18
    Novara G, Artibani W, Cunico SC et al. How accurately do Solsona and European Association of Urology risk groups predict for risk of lymph node metastases in patients with squamous cell carcinoma of the penis? Urology 2008; 71: 32833
  • 19
    Lopes A, Hidalgo GS, Kowalski LP, Torloni H, Rossi BM, Fonseca FP. Prognostic factors in carcinoma of the penis: multivariate analysis of 145 patients treated with amputation and lymphadenectomy. J Urol 1996; 156: 163742
  • 20
    Broders AC. Squamous cell epithelioma of the skin. Ann Surg 1921; 73: 14160
  • 21
    Black MM, Speer FD. Nuclear structure in cancer tissues. Surg Gynecol Obstet 1957; 165: 97102
  • 22
    Inoue T. Prognostic significance of the depth of invasion relating to nodal metastases, parametrial extension, and cell types. A study of 628 cases with Stage IB, IIA, and IIB cervical carcinoma. Cancer 1984; 54: 303542
  • 23
    Fereneczy A. Electron microscopy of histologically difficult to diagnose gynecologic neoplasms. Ultrastruct Pathol 1987; 11: 33560
  • 24
    Wilkinson EJ. Superficially invasive carcinoma of the vulva. Clin Obstet Gynecol 1991; 34: 65161
  • 25
    Anneroth G, Hansen LS, Silverman S Jr. Malignancy grading in oral squamous cell carcinoma. I. Squamous cell carcinoma of the tongue and floor of mouth: histologic grading in the clinical evaluation. J Oral Pathol 1986; 15: 1628.
  • 26
    Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 1991; 19: 40310
  • 27
    Naumann CM, Alkatout I, Al Najar A et al. Lymph-node metastases in intermediate-risk squamous cell carcinoma of the penis. BJU Int 2008; 102: 11026
  • 28
    Malhotra SM, Rouse RV, Azzi R, Reese JH. Expectant management in patients with T1 moderately differentiated penile cancer: is observation truly appropriate? J Urol 2006; 175 (Suppl.): 461