Squamous cell carcinoma antigen: a role in the early identification of nodal metastases in men with squamous cell carcinoma of the penis




To evaluate whether serum squamous cell carcinoma antigen (SCCAg) measurements may be of use in identifying nodal metastases in patients with SCC of the penis after treating the primary tumour.


The levels of SCCAg were analysed in 11 men with penile SCC between 1994 and 2001.


An elevated SCCAg level had a sensitivity of 57% (95% confidence interval, CI, 18–90%) and a specificity of 100% (CI 40–100%) for nodal metastases. Levels of SCCAg increased exponentially in patients who developed nodal metastases after treatment of the primary tumour, and were elevated before clinical or radiological evidence of nodal disease.


Either the absolute level or the rate of rise of SCCAg may be a useful tool with which to follow patients after excision of the primary tumour. It may be more sensitive than computed tomography and magnetic resonance imaging in detecting recurrence, but further evaluation is needed.


The prognosis after treating the primary lesion in squamous cell carcinoma (SCC) of the penis depends on whether there are metastases in the inguinal lymph nodes, which are found in ≈ 25% of men at presentation [1,2]. It is important to identify nodal disease as soon as possible to plan block dissection of inguinal/iliac lymph nodes before the tumour can spread to the common iliac lymph nodes, when cure is almost impossible [2]. The probability of nodal disease can partly be predicted by the stage and grade of the primary tumour, and suggests the need for either an immediate lymphadenectomy or observation [1,3,4]. In the latter group, the prediction of lymphadenopathy at an earlier stage may allow an earlier lymphadenectomy and possibly better survival.

In cervical and anal carcinoma, serum SCC-associated antigen (SCCAg) has been useful in staging and following the course of the disease [5–8]. SCCAg is a glycoprotein with relative molecular mass of 48 kDa. In the only other two published studies of its use in penile carcinoma, levels paralleled the course of disease and were higher in men with nodal metastases [9,10]. The objective of the present study was to examine the use of SCCAg in men with SCC of the penis, to determine whether it might be useful for predicting nodal metastases before clinical or radiological evidence of recurrent disease.


A consecutive series of 11 men (mean age 61 years, sd 20) with SCC of the penis were evaluated with serial SCCAg levels between 1994 and 2001. The tumour characteristics are shown in Table 1; tumours were staged using the TNM system [11] and graded by Broder's system. A patient was considered not to have nodal disease if a lymph node dissection showed no disease or there was no clinical or imaging evidence of lymph nodes for ≥ 2 years. The overall median follow-up was 3 years. Four men had localized disease, judged by clinical examination, CT of the pelvic lymph nodes and prolonged follow-up of > 2 years with no evidence of nodal disease.

Table 1. The patients’ characteristics
PatientStageGradeNodeLNDSCCAg (ng/L)RiskComment
  1. LND lymph node dissection; Risk refers to the risk of nodal metastases categorized according to a prospective validated model [1].

 1111Y1200LowAfter surgical excision of primary tumour
 2120No1200IntermediateAfter surgical excision of primary tumour
 3130No  600IntermediateAfter radiotherapy of primary tumour
 4120No  850IntermediateAfter surgical excision of primary tumour
 5130No  300IntermediateAfter surgical excision of primary tumour
 6121Y  600IntermediateAfter surgical excision of primary tumour
 7131Y4200IntermediateBefore node dissection (SCCAg to 400 afterward)
 8221Y  500HighAfter radiotherapy of primary tumour
 9131Y1900HighAfter surgical excision of primary
10331Y8200HighAfter first node dissection
11321Y1200HighAfter surgical excision of primary tumour

SCCAg levels were measured using a microparticle enzyme immunoassay (Abbott Laboratories Ltd., Diagnostic Division, Maidenhead, Berks, UK), for which the upper limit of the reference interval was 1500 ng/L. Each patient received definitive treatment for the primary tumour and lymph node dissection where appropriate. Data were log10 transformed to give geometric means before analysis by two-sample t-tests (for equal variances); the sensitivity and specificity were calculated to give the diagnostic abilities of the test. The 95% CI is given where appropriate.


Seven men had nodal disease histologically, either initially or after excision of the primary tumour. The mean SCCAg level after treatment of the primary lesion was 1610 ng/L in patients with nodal metastases and 590 ng/L in patients who did not develop nodal metastases (P = 0.06; Table 1). In the three patients who had no node dissection early but did so later the SCCAg levels increased rapidly before nodal disease became evident either clinically or on radiological imaging, and generally followed the course of the disease. In one patient the SCCAg levels decreased after nodal dissection, increased after disease recurred, decreased again after chemotherapy and increased until death (Fig. 1a). The course was similar in a second man (Fig. 1b). In four of the seven patients with nodal metastases, SCCAg levels were above the reference interval, compared with none of the four with no metastases, giving a sensitivity of 57% (CI 18–90%) and specificity of 100% (CI 40–100%). In this study, seven of 11 men had or went on to develop nodal disease, and an elevated SCCAg had a positive predictive value of 100% (CI 40–100%) and a negative predictive value of 57% (CI 18–90%). After lymph node dissection SCCAg levels decreased to below the upper limit of the reference interval in most patients.

Figure 1.

Sequential measurement of SCCAg level in two patients treated after penectomy. a shows the SCCAg level after circumcision, radiotherapy and cisplatin to treat the primary lesion (1), local recurrence (2) followed by amputation (3), the declining levels after amputation (4), increasing levels and the detection of nodal metastases (5), pelvic lymph node clearance (6), the development of distal metastases and death (7). b, the levels before penile amputation and lymphadenectomy (2), after penile amputation and lymphadenectomy (3), increasing levels before clinical detection (4) and death.


This study suggests that in men with SCC of the penis either elevated levels of SCCAg or continuously rising levels of SCCAg after treating the primary lesion might be used to predict the presence of nodal metastases and the need for a lymph node dissection. In a previous report [9], 10 of 11 men with metastases had an SCCAg level of > 1500 ng/L (sensitivity 91%, CI 59–100%), and almost all with no metastases had levels of < 2000 ng/L. The latter group, and a report of three Japanese patients, showed that SCCAg levels followed the course of the disease [10]. There are no other published studies reporting the use of SCCAg in men with SCC of the penis.

It is thought that 10–15% of men have nodal metastases which are not clinically evident at the time of initial presentation. Patients at risk of nodal metastases can be divided into low, intermediate or high risk by the initial stage and grade of the primary lesion, and this has been validated in a prospective study [1]. Those patients at low risk (T1G1, < 17%) do not require a node dissection and almost certainly all those at high risk (T2G2, T2G3, T3G2, T3G3) with a probability of > 80% of having nodal disease should have a node dissection. It is the intermediate group (T2G1, T3G1, T1G2, T1G3), representing about a third of all patients and of whom ≈ 35% may have nodal disease, that may gain the most benefit from this SCC marker. The probability of nodal disease after a positive test should increase sufficiently to compensate for the morbidity of an unnecessary nodal dissection. For that to be true, the test needs to increase the probability of nodal disease from 35% to 80%, and so the test should have a high specificity and positive likelihood ratio (≈ 7.5) [12]. The present data and those of Wishnow et al.[9] suggest that the use of SCCAg levels might meet these criteria. Although increasing SCCAg levels might indicate that metastases are present, it is not yet appropriate to recommend early nodal dissection from these data alone, as the CIs were wide because there were few patients. Nevertheless, the interpretation of the use of SCCAg from the present patients in the UK is consistent with those from Japan and America, which were reported several years apart. Taken together, these results indicate that SCCAg might be a useful marker. Furthermore, in those having lymph node dissection with an elevated SCCAg beforehand, there was a marked decrease in SCCAg levels afterward, consistent with the source arising from the lymph nodes.

Cooperation amongst urologists treating patients with SCC of the penis is needed to evaluate the usefulness of this marker, given the few patients generally presenting. This would require specific protocols for measuring SCCAg levels. Either the slope of the increase in SCCAg level, its doubling time or static values might be helpful. Such studies should be possible because patients are seen frequently after excision of the primary lesion (e.g. 2-monthly, as recommended in the European Association of Urology Guidelines, 2001). Patients who are suitable would be at intermediate risk of developing nodal disease, as indicated above.

Some of the patients in the present series were referred and may not represent the typical spectrum of patients presenting with SCC of the penis. Other methods exist for detecting lymph node metastases that could render measuring SCCAg levels unnecessary [3], including the detection of 99mTc-labelled sulphur colloid in involved lymph nodes with a γ-camera before excising the primary lesion [13,14]. The colloid is injected at the site of primary penile carcinoma 1 h before surgery. Such studies are promising and appear to have high specificity, but a longer follow-up in more patients is necessary to determine how reliable they are at excluding the eventual development of nodal metastases. Recently, positron emission tomography has also been shown to detect metastases from penile carcinoma [15,16].

In conclusion, SCCAg is the first serum marker that might have a clinical use in managing men with SCC of the penis. In some men, sequential monitoring of SCCAg might indicate developing metastases before they are detected clinically or on imaging, giving the opportunity for an early and potentially curative lymph node dissection. Levels of SCCAg might also be useful in following the response to treatment, be it surgery or chemotherapy. Further collaborative work with formal study protocols is necessary to determine its precise role in managing men with penile cancer.


SCCAg, squamous cell carcinoma, antigen.