Evaluation of the Contribution of Repeat Total Body Photography for the Early Diagnosis of Melanoma

Abstract

Introduction

Early detection of cutaneous melanoma is crucial, as prognosis is strongly determined by Breslow thickness. Total body photography (TBP) is increasingly used in high-risk populations, yet optimal surveillance strategies remain undefined. In our study, we aimed to evaluate how different follow-up approaches, including those incorporating TBP at varying frequencies, influence the early diagnosis of melanoma, as measured by Breslow thickness.

Methods

This retrospective cohort study included adult patients diagnosed with primary cutaneous melanoma at Rabin Medical Center between 2020 and 2025. Melanomas were classified into four groups according to the clinical setting at diagnosis: non-high-risk clinics, first visit to a high-risk clinic before TBP, high-risk follow-up with prior TBP only, and high-risk follow-up with both prior and repeat TBP on the day of diagnosis. Demographic, clinical, and histopathologic data were extracted. Differences in Breslow thickness were analyzed using analysis of covariance, and invasive versus in situ melanoma rates were assessed using multivariable logistic regression.

Results

A total of 282 melanomas from 249 patients were analyzed. Melanomas diagnosed in non-high-risk clinics were significantly thicker than those diagnosed in high-risk clinics (adjusted mean Breslow difference: 0.78 mm). Surveillance strategy was associated with invasiveness: invasive melanoma rates were highest in non-high-risk clinics (52%), followed by first visit to high-risk clinic (39%), patients with prior TBP only (34%), and lowest in patients diagnosed during repeat TBP (20%; overall model p = 0.03). Overall Breslow thickness did not differ significantly between melanomas diagnosed during visits with prior TBP alone and those diagnosed during repeat TBP.

Conclusion

High-risk melanoma clinics incorporating TBP are associated with earlier melanoma detection and lower rates of invasive disease. Repeated TBP was associated with lower rate of invasive melanoma compared with baseline TBP and manual comparison on the day of diagnosis.

Plain Language Summary

Melanoma is a serious skin cancer that is most treatable when found early. Doctors track its progress using “Breslow thickness,” which is a measurement of how deep the cancer has grown into the layers of the skin. Thinner cancers are less likely to spread to other parts of the body. To catch these cancers early, some patients at high risk are monitored with “total body photography,” which involves taking a full set of standardized photos of the skin to help doctors spot new or changing spots over time. While this method is common, experts are not sure how often these photo sets should be updated or if taking new photos frequently provides a real advantage over just using the original ones. This study looked at 5 years of cases to compare cancers found in general clinics against those found in high-risk clinics using different photo update schedules. The results showed that patients in high-risk clinics had their cancers caught earlier and were less likely to have “invasive” melanoma – cancer that has started to spread deeper into the skin. While updating the photos helped find these very early noninvasive spots, it did not significantly change the average thickness of the deeper tumors found. This suggests that while photography is a powerful tool for early detection, frequently repeating the process may mostly help find surface-level changes rather than preventing deeper tumor growth.

Introduction

Melanoma accounts for only 4% of skin cancer cases; however, it is responsible for 75% of skin cancer-related deaths [1]. The most important single prognostic factor for cutaneous melanoma is the Breslow classification system (i.e., Breslow’s thickness), which describes the histological thickness of the tumor [2–4]. While early-stage melanoma has an excellent 5-year survival rate of over 97%, late-stage melanoma has a significantly poorer prognosis [5]. To improve detection and diagnosis of early-stage melanoma, noninvasive imaging techniques are being used more frequently, which include total body photography (TBP). TBP offers a baseline mapping of all skin lesions, aiding in the accurate and swift identification of changing and new lesions as opposed to stable lesions, by providing a reference for tracking changes [6]. Previous studies have shown that TBP use can reduce the rate of unnecessary biopsies of benign nevi [7] and detect thinner melanomas when compared to dermatoscopic or physical examination alone [8–10].

However, there is no standardized protocol for using TBP, leading to varying practices across institutions. For example, some use TBP at baseline and combine it with physical examination and manual comparison at follow-up visits, while others may use TBP at baseline with repeated TBP at each visit, utilizing either manual or artificial intelligence-based comparisons. In some cases, combination of these approaches is also used. To date, no study has demonstrated a clear advantage of one follow-up protocol over another. This study sought to evaluate how different follow-up approaches, including those incorporating TBP at varying frequencies, influence the early diagnosis of melanoma, as measured by Breslow thickness.

Methods

At Rabin Medical Center (RMC), patients undergo skin examinations in general dermatology and plastic surgery clinics. Patients identified as being at higher risk for skin cancer are referred to specialized high-risk clinics, where TBP is incorporated into their follow-up using the ATBM and ATBM master systems (FotoFinder, Germany). These systems allow for side-by-side comparison of TBP maps, equipped with artificial intelligence software that marks new and changing lesions, and therefore enhances the ability to detect smaller changes even in patients with many lesions.

The follow-up protocol at the high-risk clinics consists of visits with a dermatologist specializing in skin cancer screening approximately every 6 months, with repeat TBP scheduled approximately every 12 months, resulting in an alternating follow-up scheme in which visits including TBP are interspersed with visits without TBP. However, because TBP is not covered by the public healthcare system, some patients discontinue the follow-up protocol after their initial high-risk clinic visit, while others defer repeat TBP examinations. Consequently, high-risk clinic visits without TBP are more frequent than visits that include repeat TBP.

The inclusion criteria for the study were adult patients diagnosed histopathologically with primary cutaneous melanoma at Rabin Medical Center between 2020 and 2025. The exclusion criteria included patients with non-primary melanoma, metastatic or recurrent melanoma, those who were lost to follow-up, and those with incomplete or unavailable variable data. To enable meaningful comparison between different follow-up settings and TBP utilization patterns, melanoma cases were subsequently classified into four groups according to the clinic and the type of visit at which they were diagnosed.

• 1.Non-high-risk group – melanoma cases diagnosed in non-high-risk clinics (e.g., general dermatology or plastic surgery clinics).
• 2.First visit group – melanomas that were diagnosed during the patient’s first visit to a high-risk clinic (dermatologists specializing in skin cancer screening), before TBP.
• 3.Prior TBP group – melanomas diagnosed during a patient’s follow-up visit at a high-risk clinic who had a previous TBP but did not undergo a repeat TBP on the day of melanoma diagnosis.
• 4.Repeat TBP group – melanomas diagnosed during a patient’s follow-up visit at a high-risk clinic who had both a previous TBP and a repeat TBP performed on the day of melanoma diagnosis.

Data Collection

RMC medical records were reviewed to extract the following data: age at diagnosis, sex, medical history (including personal and family history of melanoma and other non-melanotic skin cancers), number of nevi excised prior to diagnosis, and personal history of radiation exposure, sun exposure, and sunburn. For each melanoma diagnosed, data were collected on melanoma type, Breslow thickness, anatomical location, date of lesion detection on examination and date of biopsy. Follow-up data included the date of the first clinic visit, the visit immediately preceding melanoma diagnosis, and the last follow-up visit on or after diagnosis, along with the type of clinic attended and the expertise of the diagnosing physician, as well as dates of all TBPs prior to melanoma diagnosis.

Statistical Analysis

Baseline characteristics were summarized using descriptive statistics. Analysis of covariance with adjustments for covariates was used to compare differences in Breslow thickness between follow-up groups; Bonferroni-adjusted pairwise comparisons of estimated marginal means followed when appropriate. Differences in the percentage of invasive melanoma between follow-up groups were assessed using binary logistic regression with adjustment for covariates. To ensure the independence of observations, a sensitivity analysis was performed by excluding subsequent melanomas from repeat patients. p < 0.05 was considered to indicate statistical significance. The data were analyzed using SPSS, version 26.0 for Windows (SPSS, Inc.).

Results

For this study, 282 melanoma cases from 249 melanoma patients were included in the analysis. Out of these cases, 116 (41%) belonged to female patients, distributed similarly throughout the different groups. Cases in the non-high-risk clinics group (group 1) were of older patients (67.5 years ± 15.9 years) compared with cases from the high-risk clinic groups (groups 2–4; mean 61.5 y ± 14.6 years, p < 0.001). The TBP groups (groups 3–4) had a much higher rate of patients with personal and familial history of melanoma compared to the first visit group (group 2) and the non-high-risk group (group 1, Table 1).

Table 1.

Demographic and patient characteristics of melanoma cases by follow-up group

Follow-up group	Group 1 (N = 141)	Group 2 (N = 54)	Group 3 (N = 62)	Group 4 (N = 25)
Age, mean±SD	67.5±15.9	62.4±13.2	60.1±14.8	63.3±16.9
Female, N (Col %)	62 (44)	21 (38.9)	24 (38.7)	9 (36)
Personal Hx skin cancer, N (Col %)
Melanoma only	13 (9.2)	9 (16.7)	27 (43.5)	11 (44)
NMSC only	24 (17)	4 (7.4)	4 (6.5)	2 (8)
NMSC and melanoma	4 (2.8)	7 (13)	21 (33.9)	9 (36)
Family Hx melanoma, N (Col %)	15 (10.6)	11 (20.4)	18 (29)	10 (40)
Number of previously excised nevi, mean±SD	1.3±0.7	2.2±1.1	2.5±1.3	2.8±1.4
History of 2nd degree sunburn, N (Col %)	3 (7.5)	12 (28.6)	12 (25.5)	9 (52.9)
Time between discovery to biopsy, mean±SD, days	32.7±62.5	6±14.3	10.1±30	11.8±30.9

Most of the melanomas diagnosed in this study were subtyped as superficial spreading (31.9%) or as lentigo maligna (29.1%) with very few nodular cases. Most melanomas were diagnosed on the trunk (46.8%), whereas other locations had similar rates of occurrence (18.8% head and neck, 15% upper limbs, and 17.9% lower limbs). Anatomic location had a similar distribution between the groups.

Patients that were diagnosed at non-high-risk clinics (group 1) had a significantly higher wait time from suspicious lesion discovery to surgical biopsy of their melanoma compared to patients diagnosed at the high-risk clinics (groups 2–4; 32.7 ± 62.5 vs. 8.8 ± 25.1 days, p < 0.001).

When comparing the ratio of invasive versus in situ melanomas between the different groups, adjusted to age, sex, personal history of melanoma, and time intervals between previous follow-up date to lesion detection date, and between lesion detection date to biopsy date, the logistic regression model was significant overall (p = 0.03) and showed an association between the risk for invasive melanoma and the level of follow-up. Patients diagnosed at the non-high-risk clinics (group 1) had the highest rate of invasive melanomas (52%) followed by patients diagnosed on their first visit to high-risk clinic (group 2, 39%), followed by patients with prior TBP (group 3, 34%), and finally, patients diagnosed on the day of repeating their TBP (group 4, 20%; Table 2; shown in Fig. 1). In addition, both overall Breslow thickness and Breslow thickness calculated for invasive melanomas only were significantly greater for melanomas diagnosed in non-high-risk clinics (group 1) compared with those diagnosed in high-risk clinics (groups 2–4), with an adjusted mean difference of 0.78 mm in overall Breslow thickness, adjusted to age, sex, personal history of melanoma, and time intervals between previous follow-up date to lesion detection date, and between lesion detection date to biopsy date. However, when comparing overall Breslow thickness between the prior TBP group (group 3) and the repeat TBP group (group 4), the adjusted mean difference was −0.002 mm (95% CI, −1.02 to 1.03; p = 0.999), indicating no significant difference between patients who had only a prior TBP and those who underwent repeat TBP on the day of melanoma diagnosis (Table 2; shown in Fig. 1).

Table 2.

Clinical and histological characteristics of melanoma cases by follow-up group

Follow-up group	Group 1 (N = 141)	Group 2 (N = 54)	Group 3 (N = 62)	Group 4 (N = 25)
Breslow thickness, mean±SD, mm	0.85±1.7	0.31±0.57	0.14±0.24	0.11±0.23
In situ melanoma, N (Col %)	67 (47.5)	33 (61.1)	41 (66.1)	20 (80)
Invasive melanoma, N (Col %)	74 (52.5)	21 (38.9)	21 (33.9)	5 (20)
Breslow thickness (invasive melanomas only), mean±SD	1.62±2.07	0.79±0.67	0.41±0.24	0.54±0.18
Melanoma type, N (Col %)
Superficial spreading	45 (31.9)	17 (31.5)	17 (27.4)	9 (36)
Lentigo melanoma	41 (29.1)	14 (25.9)	16 (25.8)	7 (28)
Nodular	6 (4.3)	0 (0)	0 (0)	0 (0)
Acral	5 (3.5)	0 (0)	2 (3.2)	0 (0)
Unclassified	44 (31.2)	23 (42.6)	27 (43.5)	9 (36)
Melanoma location, N (Col %)
Unknown	2 (1.4)	0 (0)	1 (1.6)	1 (4)
Head and neck	27 (19.1)	13 (24.1)	10 (16.1)	4 (16)
Upper limb	22 (15.6)	6 (11.1)	7 (11.3)	6 (24)
Lower limb	23 (16.3)	12 (22.2)	15 (24.2)	2 (8)
Trunk	67 (47.5)	23 (42.6)	29 (46.7)	12 (48)

Fig. 1.

Proportion of invasive melanomas and Breslow thickness by follow-up group. Bars represent the percentage of invasive melanomas in each diagnostic group. Lines indicate mean Breslow thickness for all melanomas (solid line) and for invasive melanomas only (dashed line).

While most subjects contributed a single melanoma to the study, 24 individuals (9.6%) were diagnosed with multiple primary melanomas. To ensure the independence of observations, a sensitivity analysis was performed by excluding subsequent melanomas from these repeat patients. The results remained consistent with the primary analysis, indicating that the inclusion of multiple lesions per patient did not bias the differences observed in Breslow depth between the four follow-up groups.

Discussion

This retrospective cohort study evaluated the association between skin cancer screening patterns and melanoma Breslow thickness at diagnosis. Our findings demonstrate that melanomas diagnosed in high-risk clinics, staffed by specially trained physicians and equipped with dedicated tools such as TBP, were more likely to be thinner and diagnosed at an in situ stage compared with melanomas diagnosed in general dermatology and plastic surgery clinics.

Addressing the primary aim of this study, we found that among patients diagnosed in high-risk clinics by dermatologists specializing in skin cancer screening, the surveillance strategy was associated with the proportion of invasive versus in situ melanomas. Specifically, patients diagnosed following repeated TBP had the lowest rate of invasive melanomas (20%), followed by patients with a prior TBP and manual comparison only (group 3, 34%), and patients diagnosed on their first visit to a high-risk clinic without TBP comparison (group 2, 39%). Importantly, this gradient was observed despite patients in the TBP-based surveillance groups representing a substantially higher risk population at baseline compared with both the first visit high-risk clinic group and patients diagnosed in non-high-risk clinics.

A similar pattern was observed when examining overall Breslow thickness, with thinner melanomas diagnosed in patients undergoing TBP-based surveillance. However, this trend was not evident when restricting the analysis to Breslow thickness among invasive melanomas only. Given the small number of invasive melanomas diagnosed in the repeat TBP group, firm conclusions regarding differences in invasive melanoma thickness between TBP strategies cannot be drawn.

These findings are consistent with previous studies demonstrating that, in high-risk populations, the use of dermoscopy and TBP facilitates earlier melanoma detection and is associated with thinner Breslow thickness at diagnosis [8, 11–15]. Notably, however, prior studies have not directly compared different TBP follow-up strategies, nor have they specifically evaluated the added value of repeated TBP relative to reliance on baseline TBP alone. Our study therefore extends the existing literature by examining the impact of distinct TBP surveillance approaches within a real-world high-risk clinic setting.

Although the proportion of in situ melanomas was associated with surveillance intensity, and repeated TBP was linked to a higher rate of in situ diagnoses, the clinical significance of this finding warrants careful consideration. When comparing overall Breslow thickness between the TBP groups, the observed differences were neither statistically nor clinically meaningful (mean Breslow thickness of 0.14 mm in the prior TBP group versus 0.11 mm in the repeat TBP group). While the detection of in situ melanoma may confer benefits, including theoretically lower metastatic risk and less extensive surgical management in accordance with current guidelines [16], concerns regarding potential overdiagnosis have been raised in the literature [17]. Given that TBP is time-consuming and often associated with additional costs, the balance between clinical benefit and resource utilization must be considered. An alternating surveillance strategy, with repeat TBP performed every two or more follow-up visits, may represent a more cost-efficient approach, though this requires further prospective evaluation.

An additional difference between the non-high-risk group and the high-risk clinic groups was the time interval between lesion identification and biopsy. This shorter interval in high-risk clinics likely reflects greater efficiency in managing high-risk patients and may have contributed to the greater Breslow thickness observed among melanomas diagnosed in non-high-risk clinics.

Another notable observation relates to melanoma subtype distribution. The cohort consisted almost exclusively of superficial spreading and lentigo maligna melanomas, with very few nodular or acral melanomas. Given that nodular melanomas have been reported to be more common than lentigo maligna and are characterized by rapid vertical growth [18], it is possible that such tumors are less likely to be detected through scheduled TBP-based surveillance at 6–12 month intervals or that they originate in the dermis and therefore can only be diagnosed at a later stage [19]. This raises the possibility that current screening paradigms may be less effective for rapidly growing melanoma subtypes and that patient education combined with available dermatology consults are an important part of skin cancer screening, an issue that warrants further investigation.

This study has several limitations. Its retrospective, single-center design limits generalizability, and the number of patients in each high-risk subgroup was relatively small, particularly in group four and in invasive melanomas, which may have hindered the ability to reach meaningful differences. In addition, the surveillance groups were not fully comparable in baseline characteristics and in management pathways, and although multivariable analyses were used, residual confounding may persist. Despite these limitations, the observed trends were consistent and biologically plausible.

Conclusion

Our findings support the value of high-risk melanoma clinics and TBP-based surveillance in facilitating earlier melanoma detection and reducing the proportion of invasive disease at diagnosis. While visits with repeated TBP were associated with a higher rate of in situ melanoma detection, its incremental benefit over visits with baseline TBP and manual comparison, in terms of Breslow thickness, appears modest. These results suggest that TBP-based surveillance strategies may be further refined to maximize clinical benefit while maintaining efficiency. Future prospective, multicenter studies should focus on defining personalized surveillance intervals, integrating individual risk profiles and melanoma subtype characteristics, and evaluating long-term clinical outcomes. Such efforts may enable the development of optimized, evidence-based screening protocols that preserve the advantages of early detection while ensuring sustainable and high-value melanoma care.

Statement of Ethics

This retrospective cohort study was approved by the Rabin Medical Center Ethical Committee (RMC-0048-21), which also granted the study an exemption from requiring written informed consent.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

A. Zoabi: data curation, interpretation, and writing and drafting the original manuscript. I. Shavit: data curation and interpretation. T. Mimouni: methodology, statistical analysis, and interpretation. H. Feuerman: interpretation and critical revision. O. Reiter: supervision, conceptualization, interpretation, and critical revision.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The data that were collected and used for this study and that support its findings are not publicly available due to privacy reasons but may be obtained from the corresponding author upon reasonable request.