Improving the Performance of Lymph Node Staging 4621 Positron Emission Tomography PET

Squamous cancers usually have a high glycolytic rate and show increased uptake of glucose and an analogue, fluorodeoxyglucose,74 which can be labeled with 1 8F. Modern PET scanners usually also have a CT capability, to enable accurate coreg-istration of functional and anatomical information (vital when assessing nodes), which has been shown to increase the performance compared to the two tests alone75 76 (Fig. 4.16).

The first study of PET in penile cancer included ten patients with primary disease and three with recurrence, the majority of whom were not suspected to have lymphadenopathy clinically. Analyzed by patient, 4/5 of those with positive nodes were correctly identified, and there were no false positives. Because uptake of squamous carcinoma is often intense, malignant nodes measuring 0.7-1.1 cm were correctly identified as positive, but one 5 mm node was missed. Most of the primary lesions in the penis also showed uptake.77

This early study was encouraging but PET cannot overcome the fundamental problem of resolution: a recent study examining 42 clinically negative groins in 24 patients found that only one of five patients with clinically occult metastases was correctly identified, with the missed foci of nodal tumor <1 cm in each case.78

PET may have more of a role in the detection of pelvic metastases in patients with histologically or cytologically proven involved groin nodes, either at presentation or during follow-up. In a recent series of 18 patients, with analysis and surgical correlation possible in 28 hemi-pelvises, PET identified 10/11 tumor-positive cases and correctly identified all 17 negative cases™ In five patients, there were distant sites of uptake thought to represent metastasis, confirmed histologically in four. At the level of individual nodes, there was one false negative, containing a 5 mm deposit, but it is important to note that the true positive deposits were 9 mm or more in diameter,79 and that most would be expected to be detected by size criteria, although the short axis size of the lymph nodes was not specifically stated in this study.

4.6.2.2 MR Lymphography

Although MRI usually has good soft tissue contrast, tumor does not usually have a markedly different signal to lymph nodes.72 We have already mentioned that size criteria are ineffective, with in some studies well over half of involved nodes having a size smaller than the 1 cm size criterion that results in acceptable specificity.70 Morphologic criteria such as irregular border or mixed signal intensity may have a high specificity but the problem remains that these properties are easier to see in large nodes and may miss small deposits of tumor.64,80,81

The use of superparamagnetic iron oxide nanoparticles such as ferumox-tran-10 (Sinerem - Guerbet, Aulnay-sous-Bois, France) can markedly increase the contrast between tumor and normal lymph node. If injected intravenously the particles are taken up throughout the body by macrophages that accumulate in benign lymph nodes. The iron oxide results in reduced signal on T2 and T2* images, so that tumor involvement in nodes is seen as an area of relatively high signal that has failed to darken (Fig. 4.17). With this technique we have the ability to detect deposits down to a diameter of around 2 mm,82 and the results are a clear improvement on using size criteria alone. In the study published in the New England Journal of Medicine, sensitivity (specificity) for the detection of nodal involvement in prostate cancer improved from 35% (90%) to 91% (98%) with Sinerem,52 and in one study in a smaller number of patients with penile cancer, involving seven patients, sensitivity by node increased from 13% (76%) to 100% (97%).72

Fig. 4.17 Image (a) shows two deep inguinal nodes of similar size on a standard T2-weighted axial sequence. Both are of high signal. After Sinerem - image (b), the left node (arrow) becomes uniformly dark, indicating functioning (benign) lymphatic tissue. The right-sided node (arrowhead) remains partly high signal, indicating infiltration by tumor. By kind permission of Drs. S. McDougall and M. Harisinghani

Sinerem is well-tolerated, with one case of fatal anaphylactic shock with administration of the undiluted product,83 but although it remains under active study in Europe and the USA, it has not been fully licensed in either, and whether it ever will be is uncertain.

4.6.2.3 FNA Biopsy

A study of 118 patients with penile cancer published in 1991 found that 'the classification of regional nodes by clinical examination is hardly improved by additional imaging studies', with fine needle aspiration detecting no additional cancers.84 However, later studies are more encouraging. In a series of 28 patients, those with clinically suspicious groin nodes were subjected to FNA and then block dissection; 17 were histologically positive and FNA had sensitivity and specificity of 100%. However, the authors cautioned that all but one patient had stage II or III disease, and that the technique was much more difficult in impalpable nodes.85 A later study in 16 men with palpable nodes showed a sensitivity of 93% and specificity of 91% in 25 FNA samples (14 nodes were positive) performed by urologists without ultrasound guidance, giving an accuracy of 92%. In contrast, clinical assessment had an accuracy of only 66%.86

More recently a series of 43 patients with 83 clinically node-negative inguinal regions underwent ultrasound-guided FNA before either sentinel node biopsy or inguinal block dissection. Thirty four groins had suspicious nodes (length/width < 2, wide cortex or narrow/absent hilum) and underwent FNA. The sensitivity was only 39%, with a specificity of 100%, and the authors propose it as a useful technique to lessen the need for sentinel node biopsy: if positive on FNA, they proceeded straight to inguinal node dissection.37 It is clear from this study, however, that FNA, even guided by ultrasound, cannot be used to exclude tumor in groins with impalpable disease. Recent work by another group in 61 clinically node-negative patients proceeding to sentinel node biopsy on the same day showed positive cytology at ultrasound-guided FNA in eight inguinal 'basins', but false-negative results in six groins. The sensitivity of the technique is given at 74%, but it is hard to draw this conclusion from the data.88

4.6.2.4 Sentinel Node Biopsy

A modified inguinal lymph node dissection is likely to sample the first draining node of a penile cancer, but although it has less morbidity than a radical groin dissection, significant complications remain, with lymphoedema, wound necrosis and seroma often seen.89 Might it be possible to sample the first draining node and further minimize morbidity?

In a study in 1977, initially using lymphangiography, Cabanas identified a sentinel lymph node at the anterior or medial aspect of the superficial epigastric vein, medial to and above the epigastric-saphenous junction.90 In 15 patients positive for metastatic disease, this node was always involved, and sometimes the only positive node. Others describe 'the most medial inguinal node of the horizontal chain', just lateral to the pubic tubercle, as frequently enlarged,85 but there are several reports of patients with negative sentinel nodes (according to the Cabanas definition) who went on to develop pelvic nodal tumor,91,92 and it is clear that an anatomical approach alone is insufficiently sensitive to locate the 'sentinel' (or first draining) node. Subsequent work from the MD Anderson Cancer center suggested that even an 'extended' sentinel node dissection might miss a quarter of involved groins and could not be recommended.93

Lymphoscintigraphy, with injection of a radiolabeled tracer into the primary tumor and imaging of draining nodes with a gamma probe was described 30 years ago94 and node mapping with injection of blue dye was described for melanoma in 1992.95 In the penis it was first described in 2000 by Steinbecker96 and Han,97 who used both 99mtecnetium-labeled sulphur colloid and blue dye, locating the sentinel node with a g probe.

The first study to use sentinel node biopsy in penile cancer examined 90 patients with clinically node-negative disease.9 8 Lymphoscintigraphy with 9 9mTecnetium-labeled nanocolloid showed 217 sentinel nodes on imaging with a gamma camera; the following day 1 mL of patent blue dye was injected intradermally around the tumor, and the combination of blue staining and positivity with a g probe identified

Fig. 4.18 Sentinel node identification. Static image taken 2 h after injection of labeled colloid radiotracer into the skin around a penile lesion (black arrow). Uptake has only occurred in right-sided nodes, and a reinjection should be considered. The first draining node (asterisk, arrow head) is the sentinel node; the more superior node (arrow head) lies along the distal external iliac chain

Fig. 4.18 Sentinel node identification. Static image taken 2 h after injection of labeled colloid radiotracer into the skin around a penile lesion (black arrow). Uptake has only occurred in right-sided nodes, and a reinjection should be considered. The first draining node (asterisk, arrow head) is the sentinel node; the more superior node (arrow head) lies along the distal external iliac chain

208 sentinel nodes in 149 inguinal regions, with lymphatic drainage to both groins occurring in 81% of patients and an overall surgical node identification rate of 98%. Around one quarter of nodes were only positive with the gamma probe; three quarters also stained with blue dye. Sentinel nodes were positive in 19 inguinal regions, but a confirmatory regional lymph node dissection was performed only if the sentinel node was positive. At median follow-up of 36 months, there were five regional recurrences, giving a sensitivity (which given the follow-up, may be an overestimate) of around 80% (Fig. 4.18).

A similar technique was used in a further study of 140 patients with clinically node-negative groins. At least one sentinel node was seen in all but two patients, and was positive in 37 inguinal regions of 31 patients, in which case a 'standard' groin node dissection was performed." It was the only involved node in 78% of patients. The median follow-up was 52 months and the sensitivity 84%. A follow-up study by the same group showed improved results (sensitivity 95%) with some modifications to the technique: preoperative ultrasound with FNA of suspicious nodes, serial sectioning of the sentinel node, and surgical exploration of groins with nonvisualized sentinel nodes.100 A recent study of 60 patients from Denmark with clinically negative groins (or negative FNA with palpable nodes) showed a comparable sensitivity of 91%.101

Series with confirmatory lymph node dissections rather than clinical follow-up are smaller. One, using a similar technique to that described above, found metastases in 5 of 16 patients, bilateral in 3. 1 02 Again, three quarters of nodes also showed dye uptake, and all groins negative at sentinel node biopsy were negative at subsequent inguinal dissection. However, in one groin a sentinel node was not seen and tumor was subsequently found in two nodes at inguinal dissection: the sensitivity was 88%. A further series of 31 patients again had one patient with a positive node at inguinal dissection without an identifiable sentinel node, and one false-negative sentinel node, out of a total of seven positive groins, giving a sensitivity of 71%.

The findings with palpable nodes were less encouraging in a study of15 patients, nine with clinically palpable nodes, with sentinel node biopsy (without dye) followed by groin dissection. 3/18 groins without clinically suspicious nodes were positive and correctly identified using sentinel node biopsy but of the clinically suspicious groins, 7/12 of which had tumor, six were not identified by the sentinel node technique. 1 03 Importantly, many nodes at the dissection which were highly suspicious for tumor (and proved positive) did not take up tracer: the poor performance in palpable nodes is likely due to partial necrosis and lymphatic blockage.103 A similar effect has been seen in breast cancer.104

In summary, sentinel node biopsy may have a high sensitivity when no groin nodes are palpable, but there are several compelling reasons for first performing ultrasound +/- FNA. First, ultrasound may detect over half (8/14 in one study88) of involved groins, indicating the need for a formal inguinal node dissection. Second, FNA may sometimes detect disease where sentinel node biopsy does not: first, a sentinel node is not always identified,100 or may be misidentified as a 'neo' sentinel node because of rerouting of lymph drainage around a positive node obstructed by tumor.100 Second, histological analysis is inevitably prone to a small degree of error: in the study by Crawshaw et al. 2/17 positive nodes were called negative at initial histological analysis, with micrometastases found on pathological re-examination.8 8 Others have also described missed micrometastases at initial analysis, and because of the exacting demands on the pathologist do not routinely use frozen section analysis.100

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment