Successful sentinel node (SN) biopsy requires accurate knowledge of lymphatic drainage from the primary melanoma site. This can be obtained by lymphoscintigraphy. The use of the method has provided new insight into cutaneous lymphatic anatomy and function, and called attention to the massive variation in the lymphatic drainage patterns between patients. Patterns of lymphatic drainage from the skin are not clinically predictable [1–6], and it has been shown that a significant number of melanoma patients have recurrence of the disease in spite of negative SN biopsy .
Lymphatic drainage may go to multiple nodal fields and cross the midline of the body and is shown to be especially variable in the head–neck region and on the trunk [8,9]. Although most melanomas show drainage to standard regional nodal basins (axillary, inguinal and cervical regions), some patients also have drainage to lymph nodes outside these regions, usually shown with a frequency between 3 and 10% [5,10–15]. These nodes are often known as in-transit, interval or aberrant nodes and are defined as nodes lying along a lymphatic collecting vessel usually in subcutaneous fat between the melanoma site and the draining lymph node basin [5,16]. The most common locations of in-transit nodes (ITNs) are the scapular region and the flank for truncal melanomas [1,5,14] and the occipital region for melanomas located to the head–neck region [1,9].
It is well known that the metastatic status of the SN(s) is the most important prognostic factor in melanoma [17,18]. Micrometastatic disease can be present in any SN regardless of its location; therefore, also in ITNs where the frequency of metastatic involvement is shown to be nearly as high as for SNs in standard regional basins . If the SN technique is to give reliable information for correct staging, it seems important to detect and remove all SNs, including ITNs. To our knowledge, however, no larger studies have investigated the prognostic role of in-transit SNs.
The purpose of this investigation was therefore to analyse the distribution at a preoperative lymphoscintigraphy of SNs including ITNs, and their relationship with the localization of later recurrence and prognosis. This was done in a consecutive group of patients with melanoma at the trunk or in the head–neck region over a 12-year period at our hospital in Denmark.
The original group includes 975 consecutive patients, hospitalized for wide excision of a primary melanoma between January 1984 and July 1996, and referred to lymphoscintigraphy as part of a prospective protocol with the purpose of investigating the relationship between the lymphatic drainage pattern at the time of primary diagnosis and the pattern of possible later recurrence of melanoma. Inclusion criteria were a primary cutaneous melanoma located at the trunk or head and neck region, and clinically localized disease. Twenty-eight patients were excluded because of no lymph node visualization at the preoperative lymphoscintigraphy, and 36 patients were excluded because of lacking parameters, so the total group comprised 911 patients. In the recurrence and mortality analysis 357 patients were further excluded because they had tumour thickness of less than or equal to 1 mm and no evidence of ulceration or Clark levels IV–V, that is, they did not fulfil the SN inclusion criteria of today. The SN group comprises 554 patients; 353 men and 201 women, mean age of 56 years (range 22–91 years). Follow-up survival data were available for all patients and follow-up time was a median of 12 years (range 10–21 years). Demographic and scintigraphic data of the SN group are shown in Table 1. Informed consent was obtained from all patients before inclusion in the investigation. The investigation was approved by The Danish Data Protection Agency and meets the guidelines of their responsible governmental agency.
All patients underwent static lymphoscintigraphy with intradermal injection in two to four depots of radiolabelled colloid around the primary melanoma scar on the day before wide excision of the melanoma. The radiotracer used was rhenium (99m-technetium labelled rhenium sulphide colloid, CIS, Elektra-Box, Gif-Sur-Yvetts, Cedex, France) with a particle size of 4–8 nm. Injected activity was a median 20 MBq (range 1–84) in a volume of 0.05–0.2 ml. Gamma camera imaging was performed 1–5 h after the injection. Images were acquired with a General Electric STAR 300 (General Electric, Milwaukee, Wisconsin, USA) single-headed γ-camera in anterior-posterior views over the trunk and head–neck region, respectively, depending on the localization of the primary melanoma. The duration of image acquisition was 5–10 min/planar image. No body contour was shown. The tracer was injected and images described by the nuclear medicine physician on duty. Lymph nodes were in no case removed at the time of scintigraphy. In case of later clinically detectable lymph node metastases these were removed, and if located in a regional lymph node (RN) basin a complete lymph node dissection was performed.
The comparability of the groups with and without ITNs regarding age, melanoma thickness, Clark level and number of drainage regions visualized at scintigraphy was tested with Wilcoxon's unpaired test, whereas sex, melanoma type and ulceration were tested with the χ2 test. The comparisons of time to the first recurrence and survival time of the patients with and without ITNs at scintigraphy in the SN group were based on the log-rank test. The Cox regression hazard model was used to calculate the hazard ratios for a ‘hard event’ (recurrence and/or death of melanoma). A P value of less than 0.05 was considered significant (two-sided). The analyses were performed in SPSS version 13.0 (SPSS, Denmark), and SAS (SAS, Denmark).
Total and SN groups
In the total group 55 ITNs were present in 50 (5.5%) of the patients, and in the SN group (stage Ia patients excluded) 38 ITNs were present in 34 (6.1%) of the patients. In the SN group, the frequency of drainage to ITNs was around 6% for both truncal and head–neck melanomas. Eight of the patients (1% of the whole SN group) had only ITNs visualized at scintigraphy. Four patients had drainage to two ITNs, and one patient to three ITNs. The most frequent locations of ITNs were in the triangular intermuscular space and around the scapula (Table 2). The most frequent tumour localization with drainage to ITNs was the back of the chest (56%). For the remaining 94% of patients drainage to only standard RNs was observed.
Comparison of patients regarding sex, age and characteristics of the melanoma revealed no significant differences between patients with and without ITN. In the ITN group there was an overweight of women, the melanomas were slightly thicker, and ulceration more frequent compared with the RN patients (Table 1). Furthermore, a significantly higher number of drainage regions was visualized at scintigraphy in the ITN patients compared with the RN patients in both the total and the SN groups; median 2 versus 1 nodes (P<0.0001).
Time to recurrence and survival
Table 3 compares the recurrence and survival of the ITN and RN patients in the SN group, that is, patients fulfilling the present international criteria of SN biopsy (melanomas with ulceration, Clark level IV–V or >1 mm in thickness). Regarding recurrence and melanoma-related death, the frequencies were higher and the times from melanoma excision and scintigraphy to first recurrence, and to melanoma-related death, respectively, shorter in the ITN patients compared with RN patients; however, neither difference was significant (Table 3, Figs 1 and 2). The 5-year overall survival was significantly lower among the ITN patients compared with the RN patients; 61 and 75%, respectively (P=0.05), whereas there was a nonsignificant tendency toward shorter disease-free 5-year survival in the ITN group compared with the RN group (Table 3). Comparison of time from melanoma excision and scintigraphy to a ‘hard event’ (recurrence and/or death of melanoma) showed a significantly higher hazard of 1.6 (P=0.05) for a ‘hard event’ to occur during follow-up in the group of ITN patients compared with the group of RN patients. The life expectancy after a recurrence was not related to the presence of ITN.
The first recurrence was observed in a lymph node in 27% (n=150) of the patients in the SN group. For 91% of these patients, the metastatic lymph node was located to a basin visualized at the preoperative scintigraphy. Half of the ITN patients had recurrence during follow-up; 12% had their first recurrence in the RN basin shown at scintigraphy in addition to ITN; however, 6% had their first recurrence in the ITN and 12% in the skin above the ITN visualized at scintigraphy. Thus, nearly one-fifth of the ITN patients had their first recurrence in the region of the ITN.
To analyse whether the prognostic differences between the ITN and the RN groups may be explained by the significantly higher number of drainage basins visualized at scintigraphy in the ITN group, we calculated the hazard ratio for a ‘hard event’ to occur for patients having one versus two or more drainage basins visualized in the whole SN group, regardless of presence of ITNs. We found no significant difference (P=0.35). We found that the ‘hard event’ tendency, however, goes up with the number of regions visualized in the (minor) ITN group, whereas it is of no importance in the (larger) RN group; the ‘hard event’ hazard is estimated to go up by a factor 1.4 per region in the presence of ITN and only by a factor 1.0 in their absence (P=0.05).
This consecutive series of patients underwent preoperative lymphoscintigraphy before wide excision of a primary, cutaneous melanoma located at the trunk or head–neck region in the years before the SN technique was introduced. ITNs were found in 6% of the patients, also in stage Ia patients. For patients fulfilling the subsequently introduced SN criteria (ulceration or Clark levels IV–V or Breslow thickness >1 mm, and clinically without metastases), we observed significantly higher odds of an earlier ‘hard event’ (recurrence and/or death of melanoma) in the ITN group compared with the RN group, and the 5-year survival was significantly lower. When analysing the risk of recurrence and death separately there were nonsignificant trends of more and earlier recurrences and melanoma-related deaths in the ITN group compared with the RN group. The explanation that the hazard ratio when comparing the risk of recurrence and/or death of melanoma was significantly different, but not when comparing the two events separately, is probably that the number of patients alive with disease in the ITN group is greater than in the RN group. It is to be noted that 18% of the patients with ITN(s) had their first recurrence in the area where the ITN, not removed, was found at scintigraphy.
Patients with ITN(s) had significantly more drainage basins visualized at lymphoscintigraphy compared with patients with only RNs. Therefore, we investigated whether this factor was an independent factor related to recurrence and survival in the SN group. No significant difference in the risk of a ‘hard event’ for patients having one versus two or more basins visualized was observed. We, however, found that the recurrence tendency goes up with the number of regions visualized in the ITN group, whereas it is of no importance in the RN group. This differential effect, albeit in itself not statistically significant, may partly explain the worsened prognosis in the ITN group compared with the RN group; ITN may represent extra drainage basins being at risk for metastasis.
We observed a frequency of ITNs comparable with other studies (3–10%) [1,5,13–15]. In contrast to this study all these studies also included patients with melanomas located at the extremities, where drainage to ITNs is known to be less frequent . The body distribution of ITNs in this study is also comparable with other studies with the scapular region and the flank being the most frequent sites of ITNs overall [1,5,14] and the occipital region and close to the ear being the most frequent regions for ITNs in head and neck melanomas [1,9].
The histological status of the SN is the most important prognostic factor for the recurrence and survival of melanoma [17,18]. Although in-transit SNs are identified infrequently, they have a risk of malignant dissemination that has been reported to be nearly as high as in standard lymph node regions (cervical, axillary and inguinal) [1,13,14,19,20]. The general opinion has therefore been that it is advisable to treat ITN with biopsy in the same fashion as SNs in standard regional basins to stage the patients correctly and possibly offer further surgery or adjuvant therapy [10–12]. These findings of local recurrences in nearly one-fifth of ITN emphasize the importance of detecting and removing the ITN to avoid later recurrence in that region.
Few data are available on the prognostic impact of drainage to ITNs. Porter et al.  show that patients with truncal melanomas and drainage to multiple basins have an increased risk for occult metastases and thereby an aggravated prognosis. ITN may represent additional basins. This is in accordance with our finding of drainage to significantly more basins in patients with ITNs and a significantly shorter time to recurrence and/or death of the disease compared with patients with only RNs. It cannot, however, be totally excluded that a few of the additional basins represent false-positive findings at the lymphoscintigraphy (lymphangioma, lymphatic lake or skin contamination), as the lymph nodes were not removed routinely in this study. It can also not be excluded that the nonsignificant tendencies toward more ulceration and slightly thicker melanomas, two well-known prognostic risk factors, in ITN patients compared with patients with only RNs  may contribute to or even explain the poorer prognosis in the ITN patients.
The McMasters study finds most of the ITNs using a hand-held γ-probe during surgery and not by preoperative lymphoscintigraphy . In case of drainage to only ITNs at scintigraphy they suggest examination with the probe of the RN basin as well, as the RNs often also contain metastatic cells. This is confirmed by our finding of the first recurrence being in a regional basin in 12% of patients with ITN(s). In case of extracapsular extension of the in-transit metastasis or suspicion of metastatic contamination of the surgical wound McMasters et al.  proposed re-excision of the ITN site. They, however, did not suggest routinely RN dissection in case of a positive ITN without drainage to the RN basin at scintigraphy.
As mentioned above, this study was conducted before the SN biopsy era and therefore the lymph nodes visualized at lymphscintigraphy were not surgically removed routinely. Lymph nodes were not removed until they became clinically detectable. Apart from that difference the diagnostic, therapeutic and follow-up procedures underwent only few, unimportant changes during the whole period and compared with the present approach. We therefore think that our observations, from this first really long-term follow-up, in patients with ITNs of a serious prognosis including a substantial risk of later recurrence in the ITN area seem highly relevant for the view on the present approach to the SN procedure.
Limitations of the study
The SNs visualized at lymphoscintigraphy were not removed at primary surgery as the study was initiated before the SN era. This, however, enabled us to investigate the natural history of the SNs visualized with a minimum follow-up time of 10 years. Furthermore, we excluded all the patients from the analysis who, according to present indications for SN biopsy, should not undergo lymphoscintigraphy.
The image quality was slightly suboptimal because of the lack of dynamic and single-photon emission computed tomography imaging, a tracer with probably less ideal particle homogeneity, older generation γ-cameras, and probably most important the display of the scintigraphic studies by photographs of screen images. These limitations, however, should not reduce the reliability of comparison between the two patient groups of ITN and RN patients. If anything, the lack of dynamic imaging could favour a lower frequency of ITNs.
In this study, we followed 911 patients with truncal and head–neck melanomas and preoperative SN scintigraphy (without SN biopsy) for a minimum of 10 years. The study shows that patients with ITN(s) fulfilling the SN criteria have at least the same risk of recurrence and death from the melanoma disease, probably even higher, as patients with lymph nodes only in standard RN basins. This may partly be related to a higher number of drainage regions. These results strongly suggest the importance of detection and biopsy of in-transit SNs, in which metastatic disease seems to develop almost as often as in SNs in standard regional basins.
The authors thank J. Hilden for valuable statistical assistance.
1. McMasters KM, Chao C, Wong SL, Wrightson WR, Ross MI, Reintgen DS, et al.; Sunbelt Melanoma
Trial Group. Interval sentinel lymph nodes in melanoma
. Arch Surg 2002; 137:543–549.
2. O'Toole GA, Hettiaratchy S, Allan R, Powell BWEM. Aberrant sentinel nodes in malignant melanoma
. Br J Plast Surg 2000; 53:415–417.
3. Statius Muller MG, Hennipman FA, Van Leeuwen PAM, Pijpers R, Vuylsteke RJCLM, Meijer S. Unpredictability of lymphatic drainage patterns in melanoma
patients. Eur J Nucl Med 2002; 29:255–261.
4. Thompson JF. The Sydney Melanoma
Unit experience of sentinel lymphadenectomy for melanoma
. Ann Surg Oncol 2001; 8:44S–46S.
5. Uren RF, Howman-Giles R, Thompson JF, McCarthy WH, Quinn MJ, Roberts JM, et al. Interval nodes. The forgotten sentinel nodes in patients with melanoma
. Arch Surg 2000; 135:1168–1172.
6. Lock-Andersen J, Rossing N, Drzewiecki KT. Preoperative cutaneous lymphoscintigraphy in malignant melanoma
. Cancer 1989; 63:77–82.
7. Caracò C, Marone U, Celentano E, Botti G, Mozzillo N. Impact of false-negative sentinel lymph node biopsy
in patients with cutaneous melanoma
. Ann Surg Oncol 2007; 14:2662–2667.
8. Uren RF, Howman-Giles RB, Shaw HM, Thompson JF, McCarthy WH. Lymphoscintigraphy in high-risk melanoma
of the trunk: predicting draining node groups, defining lymphatic channels and locating the sentinel node. J Nucl Med 1993; 34:1435–1440.
9. O'Brien CJ, Uren RF, Thompson JF, Howman-Giles RB, Petersen-Schaefer K, Shaw HM, et al. Prediction of potential metastatic sites in cutaneous head and neck melanoma
using lymphoscintigraphy. Am J Surg 1995; 170:461–466.
10. Vidal-Sicart S, Pons F, Piulachs J, Castel T, Palou J, Herranz R. Midarm sentinel lymph nodes showing surprising drainage from a malignant melanoma
in the forearm. Clin Nucl Med 1998; 23:273–274.
11. Hunt JA, Thompson JF, Uren RF, Howman-Riles R, Harman CR. Epitrochlear lymph nodes as a site of melanoma
metastasis. Ann Surg Oncol 1998; 5:248–252.
12. Thompson JF, Hunt JA, Culjak G, Uren RF, Howman-Giles R, Harman CR. Popliteal lymph node metastasis from primary cutaneous melanoma
. Eur J Surg Oncol 2000; 26:172–176.
13. Vidal-Sicart S, Pons F, Fuertes S, Vilalta A, Rull R, Puig S, et al. Is the identification of in-transit sentinel lymph nodes in malignant melanoma
patients really necessary? Eur J Nucl Med Mol Imaging 2004; 31:945–949.
14. Sumner WE III, Ross MI, Mansfield PF, Lee JE, Prieto VG, Schacherer CW, Gershenwald JE. Implications of lymphatic drainage to unusual sentinel lymph node sites in patients with primary cutaneous melanoma
. Cancer 2002; 95:354–360.
15. Roozendaal GK, De Vries JD, Van Poll D, Jansen L, Schraffordt Koops H, Nieweg OE, Kroon BB. Sentinel nodes outside lymph node basins in patients with melanoma
. Br J Surg 2001; 88:305–308.
16. Thelmo MC, Morita ET, Treseler PA, Nguyen LH, Allen RE Jr, Sagebiel RW, et al. Micrometastasis to in-transit lymph nodes from extremity and truncal malignant melanoma
. Ann Surg Oncol 2001; 8:444–448.
17. Gershenwald JE, Thompson W, Mansfield PF, Lee JE, Colome MI, Tseng CH, et al. Multi-institutional melanoma
lymphatic mapping experience: the prognostic value of sentinel lymph node status in 612 stage I or II melanoma
patients. J Clin Oncol 1999; 17:976–983.
18. Balch CM, Buzaid AC, Soong SJ, Atkins MB, Cascinelli N, Coit DG, et al. Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma
. Clin Oncol 2001; 19:3635–3648.
19. Uren RF, Howman-Giles R, Thompson JF. Patterns of lymphatic drainage from the skin in patients with melanoma
. J Nucl Med 2003; 44:570–582.
20. Matter M, Nicod Lalonde M, Allaoua M, Boubaker A, Liénard D, Gugerli O, et al. The role of interval nodes in sentinel lymph node mapping and dissection for melanoma
patients. J Nucl Med 2007; 48:1607–1613.
21. Porter GA, Ross MI, Berman RS, Lee JE, Mansfield PF, Gershenwald JE. Significance of multiple nodal basin drainage in truncal melanoma
patients undergoing sentinel lymph node biopsy
. Ann Surg Oncol 2000; 7:256–261.