Skip to main content
. Author manuscript; available in PMC: 2015 Dec 1.
Published in final edited form as: JAMA Dermatol. 2014 Dec 1;150(12):1306–1314. doi: 10.1001/jamadermatol.2014.1348

Table 3. Relationship Between Histopathologic Pigmentation and AJCC Tumor Stage for 3,325 Primary Invasive Melanomas From 2,845 Patientsa.

No. (%) of Melanomas
Pigmented Amelanotic Amelanotic vs. Pigmented Melanoma
AJCC Tumor Stageb (n=3,058) (n=267) Adjusted OR (95% CI)c Ptrendd
 T1a 1,720 (97) 54 (3) 1 [Reference]
 T1b 412 (92) 36 (8) 2.9 (1.8-4.6)
 T2a 506 (91) 53 (9) 3.5 (2.3-5.4)
 T2b 62 (79) 16 (21) 11.1 (5.8-21.2) <.001
 T3a 168 (82) 38 (18) 9.0 (5.6-14.5)
 T3b 103 (89) 13 (11) 4.6 (2.3-9.2)
 T4a 51 (64) 29 (36) 24.6 (13.6-44.4)
 T4b 36 (56) 28 (44) 29.1 (15.5-54.9)

Abbreviations: AJCC, American Joint Committee on Cancer; OR, odds ratio; CI, confidence interval.

a

Invasive single melanomas and multiple melanomas (index and previous) with missing data for stage (n = 161) were excluded. We used marginal logistic regression with an independent correlation structure implemented in generalized estimating equations to account for the clustering of melanomas for multiple primary melanoma (MPM) patients.

b

T1a, Breslow thickness ≤1.0 mm and absence of ulceration or mitoses; T1b, Breslow thickness ≤ 1.0 mm and presence of ulceration or mitoses; T2a, Breslow thickness 1.01-2.0 mm without ulceration; T2b, Breslow thickness 1.01-2.0 mm with ulceration; T3a, Breslow thickness 2.01-4.0 mm without ulceration; T3b, Breslow thickness 2.01-4.0 mm with ulceration, T4a, Breslow thickness > 4.0 mm without ulceration; T4b, Breslow thickness > 4.0 mm with ulceration.

c

Adjusted for age (<50, 50-69, >70), sex, anatomic site (trunk, head/neck, upper extremities, lower extremities), study center, and lesion status (single primary melanoma, index MPM, or previous MPM).

d

Ptrend was calculated in the generalized estimating equation model including stage as an ordinal variable.