Radiation Oncology/Melanoma/Primary

Melanoma: Management of the Primary Site

Primary Surgery

• Wide excision, with clinical margin depending on Breslow thickness:
  • In situ: 0.5 cm
  • <=1 mm: 1.0 cm
  • 1-2 mm: 1.0 - 2.0 cm
  • >2 mm: 2.0 cm
• Sentinel lymph node biopsy for lesions > 1 mm thick
• For lesions < 1 mm thick, SLN biopsy is not routine, but should be considered if LVI, mitotic rate > 0, or Clark level IV-V, or young age.

Randomized

• UK Melanoma Study Group (1993-2001) -- surgical margin 1 cm vs. 3 cm
  • Randomized. 900 patients, high risk melanoma, thickness >=2.0 mm, trunk or limbs. Excluded palms and soles. Arm 1) 1 cm excision margin vs. Arm 2) 3 cm excision margin. ELND or SLND or adjuvant therapy not allowed.
  • 2004 PMID 14973217 — "Excision margins in high-risk malignant melanoma." (Thomas JM, N Engl J Med. 2004 Feb 19;350(8):757-66.) Median F/U 5 years
    • Outcome: local recurrence 1 cm 3.3% vs. 3 cm 2.9% (NS), in-transit/nodal recurrence 32% vs. 28%. No difference in DSS or OS. Median OS after locoregional recurrence 28 months vs. 18 months (p=0.05)
    • Conclusion: Higher locoregional recurrence rate with 1 cm margins in high risk melanomas, but no difference on survival

• Intergroup Melanoma Surgical Trial (1983-89) -- surgical margin 2 cm vs 4 cm
  • Randomized. 468 patients intermediate thickness (1.0-4.0 mm) melanoma, on trunk or proximal extremity. Arm 1) 2cm surgical margin vs Arm 2) 4 cm surgical margin. Secondary randomization to observation vs ELND (see below). Also, nonrandomized 272 patients with distal extremity or H&N enrolled prospectively to 2 cm excision (Group B)
  • 2001 PMID 11258773 (Free full text) — "Long-term results of a prospective surgical trial comparing 2 cm vs. 4 cm excision margins for 740 patients with 1-4 mm melanomas." (Balch CM et al. Ann Surg Oncol. 2001 Mar;8(2):101-8.) Median F/U 10 years
    • Outcome: surgical margin 2 cm 0.4% vs. 4cm 0.9% (NS), anytime failure 2.1% vs. 2.6% (NS). 10-year DSS 70% vs. 77% (NS). LR profoundly affected by presence of ulceration (RR approx 6)
    • LR impact: LR strongly associated with survival. 5-yr OS of 9% for pts with LR as first relapse (n=13), 11% for any LR (n=28), vs 5-yr OS 86% if no LR. Higher LR for head/neck 9.4%, vs 5.3% dist ext, 3.1% trunk, 1.1% prox ext. Patient who had LR all died of melanoma within 10 years.
    • Conclusion: 2 cm margins are safe for intermediate thickness melanomas

• Swedish Melanoma Study Group (1982-1990) -- surgical margin 2 cm vs 5 cm
  • Randomized. 769 patients, intermediate thickness (0.9 - 2.0 cm). Excluded H&N, distal extremities, vulva. Arm 1) resection margin 2 cm vs. Arm 2) resection margin 5 cm
  • 1996 PMID 8646678 -- "Resection margins of 2 versus 5 cm for cutaneous malignant melanoma with a tumor thickness of 0.8 to 2.0 mm: randomized study by the Swedish Melanoma Study Group." (Ringborg U, Cancer. 1996 May 1;77(9):1809-14.) Median F/U 5.8 years
    • Outcome: Local recurrence 2 cm 0.8% vs. 5 cm 1.0% (NS) No difference in DFS, DSS, or OS
    • Conclusion: Recommend excision margin of 2 cm

• France/UK (1981-2000) -- surgical margin 2 cm vs. 5 cm
  • Randomized. 337 patients, melanoma <=2.0 cm. Excluded age >70, toe/nail/finger, and acral melanoma. Arm 1) 2 cm excision vs. Arm 2) 5 cm excision. Second randomization for adjuvant isoprinosine
  • 2003 PMID 12673721 -- "Surgical margins in cutaneous melanoma (2 cm versus 5 cm for lesions measuring less than 2.1-mm thick)." (Khayat D, Cancer. 2003 Apr 15;97(8):1941-6.) Median F/U 16 years
    • Outcome: Local recurrence 2 cm 0.6% vs. 5 cm 1.8 cm (NS). Median TTR 3.5 years. 10-year DFS 85% vs. 83% (NS). No difference for second randomization.
    • Conclusion: Surgical margin of 2 cm sufficient for melanoma <=2 mm

• WHO Melanoma Programme Trial #10 (1980-?) -- surgical margin 1 cm vs 3 cm
  • Randomized. 612 patients with clinical Stage I melanoma, thickness <=2 mm. Facial or digital melanoma excluded. Arm 1) 1 cm surgical margin vs. Arm 2) 3 cm surgical margin
  • 1991 PMID 2009058 -- "Narrow excision (1-cm margin). A safe procedure for thin cutaneous melanoma." (Veronesi U, Arch Surg. 1991 Apr;126(4):438-41.). Mean F/U 8 years
    • Outcome: Local recurrence 1cm 1.3% vs 3 cm 0% (NS). 8-year OS 1cm 90% vs. 3 cm 90% (NS); DFS 82% vs. 84% (NS). First event most frequently nodal relapse (7%).
    • Conclusion: Narrow 1cm excision as effective as wide excision

Retrospective

• Sydney Melanoma Unit; 2005 (1951-1996) PMID 15650644 -- "Histopathologic excision margin affects local recurrence rate: analysis of 2681 patients with melanomas < or =2 mm thick." (McKinnon JG, Ann Surg. 2005 Feb;241(2):326-33.)
  • Retrospective. 2681 patients, cutaneous melanoma <=2mm. SM-. Median F/U 7 years
  • Outcome: 10-year LR 2.9%, predictors for LR were margin of excision and thickness. If margin 1cm excluded, margin status no longer significant. LR margin <8mm 3.8% vs. 8-15mm 1.6%. Margin status not predictive of survival
  • Conclusion: Pathologic margin affected risk of LR, but only if <1 cm clinical. It does not affect survival

• MD Anderson; 1998 (1985-1996) PMID 9641453 -- "Surgical margins and prognostic factors in patients with thick (>4mm) primary melanoma." (Heaton KM, Ann Surg Oncol. 1998 Jun;5(4):322-8.)
  • Retrospective. 278 patients, thick primary melanoma >4 mm (median 6.0 mm), ulceration 57%. Median F/U 2.2 years
  • Outcome: LR 12%. LR or margin width not associated with DFS or OS, but nodal status, thickness, and ulceration was
  • Conclusion: 2cm margin sufficient

Desmoplastic and/or Neurotropic Melanoma

• Desmoplastic melanomas were initially believed to have a high recurrence rate (~50% in early literature review from 1995). This may have been related to high rate of neurotropism, and high frequency with H&N location, which may compromise surgical margins
• However, desmoplastic melanoma with appropriate surgical resection appears to have local recurrence rate of 4-7%.
• Neurotropic desmoplastic melanoma had historically local recurrence of 20-30%, but Michigan data suggest low recurrence rate with appropriate surgical margins
• Role of initial adjuvant RT is not clear, but may be reasonable if SM are <1cm or with neurotropism. Whether RT has any impact on development of distant disease and overall survival has not been established

• Queensland Melanoma Project, Australia; 2008 (1997-2006) PMID 18366400 -- "Desmoplastic melanoma: the role of radiotherapy in improving local control." (Foote MC, ANZ J Surg. 2008 Apr;78(4):273-6.)
  • Retrospective. 24 patients with desmoplastic melanoma, surgical excision followed by postop RT. Median thickness 5.2 mm. SM <1 cm in 71% patients
  • Outcome: 3-year in-field RFS 91%, OS 83%
  • Conclusion: Adjuvant RT may have been effective in reducing rate of local recurrence

• Sydney Melanoma Unit, Australia
  • 2008 (1996-2007) PMID 18823042 -- "Desmoplastic neurotropic melanoma: a clinicopathologic analysis of 128 cases." (Chen JY, Cancer. 2008 Nov 15;113(10):2770-8.)
    • Retrospective. 128 patients with desmoplastic neurotropic melanoma. Wide local excision. Median thickness 4 mm. SM >0.5 cm in 70%, >1 cm in 40%. Adjuvant RT in 21% (typically for thicker tumors (median thickness 7 mm), H&N locations, and close surgical margins). RT 33/6 or 50/25
    • Outcome: local recurrence surgery 6% vs surgery + RT 7%. No LR with surgery only if SM >1cm. Predictors were SM+ and H&N location. DM surgery 17% vs. surgery + RT 11%.
    • Conclusion: Local recurrence lower than historical controls. When margins compromised (possibly <1cm), RT may reduce local recurrence
  • 1998 (1979-1995) PMID 9740077 -- "Desmoplastic and desmoplastic neurotropic melanoma: experience with 280 patients." (Quinn MJ, Cancer. 1998 Sep 15;83(6):1128-35.)
    • Retrospective. 280 patients (desmoplastic melanoma 190, desmoplasti neurotropic melanoms 90). Male-female 1.7:1, median thickness 2.5 mm, amelanotic 44%
    • Outcome: 5-year OS 75%. Local recurrence 11%, regional/in-transit 12%, distnat mets 14%. Desmoplastic melanoma LR 7% vs. desmoplastic neurotropic melanoma 20% vs. historical overall melanoma 3%. As a function of thickness, comparable recurrence rate for desmoplastic melanoma. LR higher if SM <1 cm or if neurotropism present
    • Conclusion: Comparable survival between DM, DNM, and other cutaneous melanomas. Lower rate of LN mets

• Michigan; 2005 (1997-2004) PMID 16080180 -- "Wide excision without radiation for desmoplastic melanoma." (Arora A, Cancer. 2005 Oct 1;104(7):1462-7.)
  • Retrospective. 49/65 patients with cutaneous desmoplastic melanoma. Wide excision (>2 cm in 63%, >1cm in 100%), most H&N patients required reconstruction. No RT given. Mean thickness 4.2 mm, 38% had thickness >4 mm. Neurotropism in 32%. Mean F/U 3.7 years, minimum 2 years
  • Outcome: Local recurrence 4% (1 neurotropic, 1 not). LN mets no neurotropism 4% vs neurotropism 28% (SS)
  • Conclusion: Local recurrence considerably less than historically reported; careful attention to appropriate margins produces excellent local control

• University of Florida; 2005 PMID 16062083 -- "Neurotropic melanoma of the head and neck with clinical perineural invasion." (Newlin HE, Am J Clin Oncol. 2005 Aug;28(4):399-402.)
  • Retrospective. 3 patients with clinical perineural invasion. 2 patients incomplete resection, and received definitive RT 74.4 Gy BID. 1 patient postop RT after multiple recurrences of desmoplastic neurotropic melanoma.
  • Conclusion: RT may provide relatively long-term local control in patients with clinical perineural invasion

• UCLA; 2003 (1976-1997) PMID 12784232 -- "Efficacy of radiation therapy in the local control of desmoplastic malignant melanoma." (Vongtama R, Head Neck. 2003 Jun;25(6):423-8.)
  • Retrospective. 44 patients, desmoplastic melanoma, H&N 68%, neurotropism 48%. Local recurrence in 21/44 (48%), and extent of SM couldn't be determined. 14/21 local recurrences received postop RT (32% overall). Median dose 50 Gy (44-66 Gy), no regional LN coverage. Indications for RT: R2 resection, R1 resection, deliberate close SM, surgical concern about adequacy of resection, or multiply recurrent tumor
  • Outcome: Local recurrence RT 0% vs. no RT 57% (SS). DM rate comparable no RT 35% vs. RT 40% (NS). No difference in recurrence rate by neurotropism. Local control in 4/4 patients with R1/R2 disease
  • Conclusion: Postop RT effective in recurrent disease; consider RT as part of initial adjuvant treatment; consider RT as primary treatment if cannot clear margins

• Harvard; 1995 (1985-1992) PMID 7812919 -- "Desmoplastic neurotropic melanoma. A clinicopathologic analysis of 28 cases." (Carlson JA, Cancer. 1995 Jan 15;75(2):478-94.)
  • Retrospective. 28 patients, desmoplastic neurotropic melanoma. Most H&N (75%) and amelanotic (57%), associated LM in 56%. Mean thickness 4.1 mm.
  • Outcome: Local recurrence 27%, visceral mets 11%, survival 81%. Literature review composite LR 50% (22-77%)
  • Conclusion: Present at more advanced stage locally, but may be associated with better survival stage for stage

Adjuvant RT to Primary Site

• Sydney Melanoma Unit, Australia; 2000 (1989-98) - PMID 10618610 — "Locally advanced melanoma: results of postoperative hypofractionated radiation therapy." (Stevens G, Cancer. 2000 Jan 1;88(1):88-94.)
  • Retrospective. 174 patients treated with postop RT (primary n=35, or lymph nodes n=139). RT 33/6 over 18 days. Median F/U 2.5 years
    • For primary site (Group A): RT given initially (n=11) or after recurrence and re-exsision (n=21). Indications: SM+, SM close, desmoplastic histology, PNI, satellitosis, early/multiple recurrences
    • For adjuvant LN (Group B): RT given initially (n=107) or after recurrence (n=35). Indications: SM+, ECE, multiple LN+, large LNs, PNI, LVI, parotid node involvement
  • Outcome: In-field local recurrence 11%. Recurrence rate if RT given in <18 days 4% vs. >=18 days 15% (p=0.06). 5-year OS Group A 49% vs. Group B 38% (p=0.07)
  • Toxicity: symptomatic lymphedema 58%, worse if initial dissection.
  • Conclusion: Postop RT improved local control compared with published series. Adjuvant RT should be considered for high risk patients, including SM+, desmoplasia with neurotropism, multiple LN+, and large LN with ECE

• Princess Margaret; 1981 (1958-1979) PMID 7306920 -- "Radiotherapy in nonlentiginous melanoma of the head and neck." (Harwood AR, Cancer. 1981 Dec 15;48(12):2599-605.)
  • Retrospective. 37 patients with superficial spreading or nodular melanoma of H&N. Surgery R2 margin (n=6), surgery with close or R1 margin (n=15), recurrent disease (n=16). Variety of RT schedules
  • Outcome: Local control in gross disease 4/6 (67%), microscopic disease 14/15 (93%), recurrent disease 2/16 (12%)
  • Radiobiology: Control improved with higher dose/fx, especially <4 Gy 25% vs >4 Gy 71%
  • Conclusion: Nonlentiginous melanoma not radioresistant, RT deserves further study

• Institution?; 1958 PMID 13533685 -- "Malignant melanoma; a combined surgical and radiotherapeutic approach." (Dickson RJ, Am J Roentgenol Radium Ther Nucl Med. 1958 Jun;79(6):1063-70.)
  • Retrospective. 234 patients (local excision 71, radical excision 42, local excision + adjuvant RT 121). RT given with radium (brachytherapy or teleradium) or orthovoltage x-rays, dose 50/10
  • Outcome: No local control. OS comparable between radical surgery and local surgery + RT

Definitive RT

Lentigo maligna & Lentigo maligna melanoma

• Surgical excision would be considered the first line treatment
• However, in elderly patients large lesions can often be present on the face, and radiotherapy may be warranted if they have significant comorbidities
• Historically, Grenz or soft X-ray therapy was used, with success rate >90%
  • In Europe, the Miescher technique was popular, involving contact therapy (50% depth dose 1mm) and use of very high doses
• Lesions may take up to 2 years to completely regress, and careful clinical follow up is required

• Zurich; 2002 (1950-2000) PMID 12072074 — "A retrospective study of 150 patients with lentigo maligna and lentigo maligna melanoma and the efficacy of radiotherapy using Grenz or soft X-rays." (Farshad A, Br J Dermatol. 2002 Jun;146(6):1042-6.)
  • Retrospective. 150 patients (LM 96, LMM 57). 90% of lesions were on the face (cheek & nose 46%, eye & ear & temple 24%). Grenz rays (12 kV) in 64%, contact x-rays (20-30 kV). Dose LM 100/10 - 120/10 @10-12 Gy/fx at 3-4 day intervals. Dose LMM 42/6 - 54/6 @ 7-9 Gy/fx at 3-4 day intervals. Margin 7-10 mm. Mean F/U 8 years
  • Outcome: Recurrence rate 7% (LM 5%, LMM 3%); mean TTR 3.8 years. LR 5/7, LN 2/7
  • Conclusion: RT is curative for LM and LMM.

• Munich; 2000 (1987-1998) PMID 10954659 -- "Fractionated radiotherapy of lentigo maligna and lentigo maligna melanoma in 64 patients." (Schmid-Wendtner MH, J Am Acad Dermatol. 2000 Sep;43(3):477-82.)
  • Retrospective. 64 patients (LM 42, LMM 22). In LMM patients, excision of nodular part prior to RT. RT superficial x-rays, 100/10 over 2 weeks. Median F/U 1.2 years
  • Outcome: Recurrence LM 0%, LMM 9% (both slavaged with excision)
  • Toxicity: Cosmetic outcomes good or excellent
  • Conclusion: Fractionated RT effective method

• Westmead Hospital, Australia; 1996 PMID 8826745 -- "Radiotherapy for melanotic freckles." (Christie DR, Australas Radiol. 1996 Aug;40(3):331-3.)
  • Retrospective. 5 patients, 7 lentigo maligna. RT equivalent to at least 44/11
  • Outcome: No local failures
  • Toxicity: Favorable cosmetic result in all patients
  • Conclusion: RT safe and effective

• Princess Margaret
  • 1994 (1968-1988) PMID 8053696 -- "Lentigo maligna of the head and neck. Results of treatment by radiotherapy." (Tsang RW, Arch Dermatol. 1994 Aug;130(8):1008-12.)
    • Retrospective. 54 patients with LM of H&N region, treated with surgery (33%) or RT (67%)
    • Outcome: LC surgery 3-years 94% vs. RT 5-years 86%. No metastases
    • Toxicity: Poor cosmesis 11% (pallor, atrophy, or telangiectasia)
    • Conclusion: Conventional fractionated RT simple and effective; excellent alternative to surgery
  • 1983 (1958-1982) PMID 6863069 -- "Conventional fractionated radiotherapy for 51 patients with lentigo maligna and lentigo maligna melanoma." (Harwood AR, Int J Radiat Oncol Biol Phys. 1983 Jul;9(7):1019-21.)
    • Retrospective. 51 patients (LM 23, LMM 28). RT orthovoltage (100-250 kV), minimum 1cm margin. Dose 35/5, 45/10, or 50/15 depending on size of lesion (<3cm, 3-5cm, >5cm)
    • Outcome: LM local control 90%, 2 local failures, both salvaged. LMM local control 92%, 2 local failures, both salvaged; 1 regional/distant mets
    • Conclusion: RT simple and effective for this form of melanoma
  • 1982 PMID 7118548 -- "Radiation therapy for melanomas of the head and neck." (Harwood AR, Head Neck Surg. 1982 Jul-Aug;4(6):468-74.)
    • Retrospective. 77 patients with melanoma of H&N (LM 13, LMM 18)
    • Outcome: Local control LM 85%; LMM 94%
    • Conclusion: RT simple and effective treatment for LM and LMM
  • 1980 (1958-1977) PMID 7379026 -- "The radiotherapy of lentigo maligna and lentigo maligna melanoma of the head and neck." (Dancuart F, Cancer. 1980 May 1;45(9):2279-83.)
    • Retrospective. 23 cases (LM 8, LMM 15), treated with orthovoltage RT (100-280 keV). RT 35/5 to 45/10 to 50/20 to 60/25 fxs (mostly 45-50 in 10 fx)
    • Outcome: LM 1/8 (12%) recurred at margin, and salvaged with further RT. LMM 1/15 (7%) recurred centrally, and was salvaged with excision. Time to regression up to 2 years
    • Toxicity: majority excellent cosmesis
    • Conclusion: Conventional fractionated orthovoltage RT deserves further consideration

• Institution?; 1976 PMID 942213 -- "Treatment of melanotic freckle with x-rays." (Kopf AW, Arch Dermatol. 1976 Jun;112(6):801-7.)
  • Retrospective. 16 patients, lentigo maligna. Treated with Miescher technique
  • Outcome: 5/16 (31%) local recurrence or persistence; 3/16 (19%) developed metastatic melanoma
  • Conclusion: Procedure has been abandoned, pending further clarification

• Institution?; 1971 PMID 5091872 -- "Hutchinson's freckle treated by the Miescher technique utilizing a Grenz-Ray machine adapted for the purpose." (Orfuss A, Arch Dermatol. 1971 Apr;103(4):456-8.)

Acral Melanoma

• Louisian State; 1999 PMID 10474984 -- "Radiotherapy of acral lentiginous melanoma of the foot." (Harwood AR, J La State Med Soc. 1999 Jul;151(7):373-6.)
  • Case report. 4 acral lentiginous melanoma of foot. Treated by RT alone, using 0-7-21 technique (24/3)
  • Outcome: LR excellent, all patients well palliated
  • Conclusion: RT recommended if not suitable for surgery

General Melanoma

• Frankfut, Germany; 1963 (1935-1960) PMID 13953619 -- "Radiation therapy of primary and metastatic melanoma." (Hellriegel W, Ann N Y Acad Sci. 1963 Feb 15;100:131-41.)
  • Retrospective. 259 patients (N0 64%, N+ 36%). RT 6000 r in 2 weeks using 60 kV, followed in 6-8 weeks by second course 4000-5000 r in 2 weeks. Alternatively 7000-8000 r in 4 weeks with electrons 15-20 MeV
  • Outcome: 5-year OS 48%, 10-year OS 29%. If N0, 62% and 42%. If N+ 23% and 12%. By method of treatment: 5-year OS RT alone 62% vs. excision + RT 38% vs. RT + excision 86%
  • Conclusion: Malignant melanoma can be successfully treated by RT; high doses are required

Brachytherapy

• Lille, France; 2006 PMID 16508600 -- "[Interstitial brachytherapy in management of primary cutaneous melanoma: 4 cases] - [Article in French]" (Mortier L, Ann Dermatol Venereol. 2006 Feb;133(2):153-6.)
  • Retrospective. 4 cases (3 LMM, 1 nodular), thick (>3mm) melanoma, treated with interstitial brachytherapy. Median F/U 4 years
  • Outcome: No local relapse
  • Toxicity: Cosmetic and functional outcome excellent
  • Conclusion: Interstitial brachytherapy feasible for primary control, if surgery difficult or impossible

Adjuvant Interferon

• DeCOG (2001-2004) -- Interferon x1.5 years vs x5 years
  • Randomized. 840 patients, resected cutaneous melanoma, minimum thickness 1.5 mm, clinically N0 (SLN in 76%, SLN+ in 18%). Arm 1) IFN alpha2 3MU 3x/week x18 months vs Arm 2) same IFN x60 months
  • 2010 PMID 20048184 -- "Efficacy of Low-Dose Interferon {alpha}2a 18 Versus 60 Months of Treatment in Patients With Primary Melanoma of >= 1.5 mm Tumor Thickness: Results of a Randomized Phase III DeCOG Trial." (Hauschild A, J Clin Oncol. 2010 Feb 10;28(5):841-6. Epub 2010 Jan 4.) Median F/U 4.3 years
    • Outcome: Relapse-free survival 1.5 years 76% vs 5 years 73% (NS), DMFS 82% vs 80% (NS), OS 86% vs 85% (NS)
    • Conclusion: Prolongation of IFN therapy showed no clinical benefit