PRAME in Melanoma: Diagnosis, Prognosis, and Treatment

PRAME, short for Preferentially Expressed Antigen in Melanoma, is a protein that has become one of the most useful tools in skin pathology for distinguishing melanoma from benign moles, and it is now emerging as a direct therapeutic target. First identified in melanoma cells in the 1990s, PRAME belongs to a family of proteins normally active only in the testes and placenta but switched on in many cancers. Its dual role as a diagnostic stain on biopsy slides and a bull’s-eye for experimental immunotherapies makes it unusually versatile in oncology, though its behavior varies enough across melanoma subtypes and patient ages that understanding the nuances matters.

What PRAME Does in Cancer Cells

In healthy adults, PRAME is essentially silent in most tissues. When cancer cells turn it on, the protein gives them a survival edge by blocking one of the body’s built-in braking systems: signaling through retinoic acid receptors. Retinoic acid, a derivative of vitamin A, normally pushes cells toward maturing, slowing down, and eventually dying on schedule. PRAME sits on the retinoic acid receptor and prevents the signal from going through, which means the cell keeps dividing instead of differentiating or self-destructing.¹ When researchers knocked PRAME down in melanoma cells that were resistant to retinoic acid, the cells became sensitive to its growth-stopping effects again, both in lab dishes and in animal models.¹ Beyond this retinoic acid blockade, PRAME is involved in cell proliferation, resistance to programmed cell death, and metastatic behavior more broadly.²

What controls whether PRAME gets turned on in the first place? A key factor is a chemical tag on the gene called DNA methylation, which normally keeps the gene locked. When melanoma cells lose this methylation, PRAME expression ramps up. A transcription factor called MZF1 cooperates with this loss of methylation to push PRAME production even higher. In experiments, treating melanoma cells with a drug that strips away methylation, combined with adding MZF1, dramatically increased PRAME protein levels.³ This epigenetic switch helps explain why PRAME is absent in normal skin cells but abundant in melanoma, and why its expression can vary even among tumors of similar stage.

How Pathologists Use PRAME to Diagnose Melanoma

The most immediate clinical impact of PRAME is at the microscope. Pathologists regularly encounter melanocytic lesions where standard microscopy alone cannot confidently distinguish a worrisome mole from an early melanoma. PRAME staining, applied as an immunohistochemistry test on biopsy tissue, lights up the nuclei of cells that express the protein. Because melanoma cells frequently express PRAME and benign mole cells usually do not, a strong PRAME stain tips the diagnosis toward malignancy.

A meta-analysis pooling data from multiple studies found that at a threshold where at least half the lesional cells stain positive (the 3+ level), PRAME had a sensitivity of about 74% and specificity of about 92% for primary cutaneous melanoma.⁴ When the threshold was raised to 75% of cells staining positive (4+), specificity climbed above 95%, meaning very few benign lesions would be falsely called melanoma, but sensitivity dropped to around 68%.⁴ Individual studies have confirmed this tradeoff. In one diagnostic accuracy study, PRAME 4+ scores showed 63% sensitivity and 97% specificity, with excellent agreement between different pathologists reading the stain.⁵

The practical question for pathologists is where to set the bar. A strict 4+ threshold catches fewer melanomas but rarely mislabels a mole, while a looser 3+ or even 2+ threshold catches more melanomas at the cost of some false alarms. The meta-analysis suggested the 3+ threshold strikes a better overall balance for routine use.⁴ This is especially relevant for thin melanomas, where distinguishing early-stage cancer from a dysplastic mole has the highest stakes: a study comparing thin melanomas to nevi found that only about a quarter of thin melanomas reached the 4+ level, while lowering the threshold improved detection without nevi crossing into false-positive territory.⁶

PRAME in Lentigo Maligna and Surgical Margins

One area where PRAME has proven particularly valuable is lentigo maligna, a type of melanoma in situ that appears on chronically sun-damaged skin, often on the face. These lesions can be difficult to define under the microscope because the abnormal melanocytes scatter along sun-damaged skin in ways that blur the boundary between tumor and normal tissue. Getting clear surgical margins matters enormously for preventing recurrence, but standard staining sometimes makes it hard to tell where the melanoma ends.

PRAME staining addresses this problem directly. In a study of lentigo maligna biopsies and excisions, PRAME showed 93.5% sensitivity and 94.7% specificity for diagnosing lentigo maligna, and its margin measurements correlated closely with conventional methods.⁷ The stain essentially highlights the melanoma cells in a field of background sun damage, making it easier for the pathologist to define where the lesion truly stops. A separate analysis confirmed this high diagnostic accuracy while noting that occasional cases still require careful interpretation alongside standard staining.⁸

Behavior Across Melanoma Subtypes

PRAME does not perform identically across all melanoma types, and these differences affect how much a pathologist can rely on it. The strongest and most consistent staining is seen in conventional cutaneous melanomas and their metastases. In one large series, diffuse nuclear PRAME staining was found in roughly 90% of primary cutaneous melanomas and 93% of metastatic melanomas.⁹ That same study, however, found that neither of the two desmoplastic melanomas in the cohort expressed PRAME at all, a finding consistent with the general understanding that desmoplastic melanoma, a rare fibrous variant, is a blind spot for this marker.⁹

Acral lentiginous melanoma, the type found on palms, soles, and under nails, does express PRAME, though at somewhat lower rates. One study found PRAME in about 89% of acral melanomas at any level of staining, but when using the 50% positivity threshold, sensitivity dropped to around 69% while specificity against acral nevi remained 100%.¹⁰ A separate investigation reported similar findings, with sensitivity of about 53% at the 50% threshold but very high specificity of 98%.¹¹ So a positive PRAME stain in an acral lesion is strong evidence of melanoma, but a negative stain does not rule it out.

In mucosal melanomas of the head and neck, a study applying a 60% positivity cutoff correctly identified all benign mucosal melanocytic lesions as negative and about 78% of malignant tumors as positive. A notable subset of PRAME-negative mucosal melanomas clustered in the palatal area.¹² The consistent theme across subtypes is that a strong PRAME-positive result is highly informative, while a negative result must be weighed carefully alongside other pathologic features rather than taken as reassurance.

Where PRAME Staining Can Mislead

The most recognized diagnostic pitfall involves spitzoid lesions, a group of melanocytic tumors that can look alarming under the microscope but are often benign, especially in children and young adults. Both the meta-analysis and individual studies have found that PRAME performs worst in this category, with the lowest sensitivity and the lowest overall diagnostic accuracy.⁴ That said, when PRAME is diffusely positive in a spitzoid tumor, it correlates with genomic abnormalities including TERT promoter mutations, which are associated with truly malignant behavior.¹³ The recommendation from multiple groups is to use PRAME cautiously in spitzoid neoplasms, always alongside other diagnostic tools rather than as a standalone test.¹⁴

Another pitfall is that PRAME is not perfectly specific to melanocytic tumors. Some non-melanocytic cancers, including certain poorly differentiated skin tumors like atypical fibroxanthomas, can stain positive for PRAME.¹⁵ Pathologists working up an undifferentiated tumor cannot assume that PRAME positivity means melanoma without considering the broader context.

The Pediatric Problem

PRAME expression turns out to be strongly age-dependent, which creates a specific problem for diagnosing melanoma in children and teenagers. A study comparing pediatric, young adult, and adult melanomas found that only 4% of pediatric melanomas showed diffuse PRAME staining, compared to about 16% in young adults and 50% in adults.¹⁶ Absolute expression levels followed the same pattern. This means relying on PRAME to confirm a melanoma diagnosis in a child is unreliable, as the stain will be negative in most cases even when the tumor is genuinely malignant. Separate work on pediatric spitzoid tumors and atypical Spitz tumors has confirmed that PRAME is typically not expressed in these lesions, though PRAME positivity in a pediatric setting can still be informative when present.¹⁷

The age dependence also has implications beyond diagnosis. If PRAME-targeted drugs make it to wider clinical use, young patients whose tumors express little PRAME would be poor candidates for those therapies.

PRAME as a Prognostic Marker

Beyond diagnosis, there is interest in whether PRAME expression levels can predict how aggressively a melanoma will behave. The evidence here is more mixed than on the diagnostic side. In a study of 123 cutaneous melanoma cases, higher PRAME staining extent and intensity were associated with worse overall survival in preliminary analysis, but when researchers accounted for tumor thickness, only thickness remained a significant independent predictor.¹⁸ In other words, PRAME may track with more advanced disease rather than independently driving a worse outcome once you control for how thick the tumor already is. One study also found no clear correlation between diffuse PRAME expression and the common BRAF or NRAS driver mutations in cutaneous melanoma.¹⁹

The story is different in uveal melanoma, a cancer of the eye where prognostic markers are critically needed because there are limited treatment options once the tumor has spread. Uveal melanomas are classified into risk categories using a 15-gene expression profile (15-GEP), with class 2 tumors carrying a much higher metastatic risk. PRAME status adds independent prognostic information on top of this classification. In a large prospective study, patients with class 1 tumors that were PRAME-negative had a five-year metastasis-free survival of about 96%, compared to roughly 81% for class 1 tumors that were PRAME-positive. Among the higher-risk class 2 tumors, PRAME-negative cases had about 58% five-year metastasis-free survival, while PRAME-positive class 2 tumors dropped to around 45%.²⁰ Earlier work had established this association between PRAME positivity and shorter time to metastasis in class 2 uveal melanoma specifically, and also linked it to melanoma-specific death.²¹ The combination of gene expression profiling and PRAME status is now used clinically to stratify uveal melanoma patients for surveillance and enrollment in clinical trials.

PRAME-Targeted Immunotherapy

Because PRAME is abundant on melanoma cells but nearly absent from normal adult tissues, it makes an attractive target for immunotherapy. Two main strategies are in clinical development: engineered T-cell therapy and bispecific molecules that redirect a patient’s own immune cells to attack PRAME-positive tumors.

IMA203 is an autologous T-cell receptor therapy, meaning a patient’s own T cells are collected, genetically engineered to recognize the PRAME peptide displayed on tumor cell surfaces, expanded in a lab, and infused back into the patient. A phase 1 trial in patients with advanced melanoma who had already progressed on checkpoint inhibitors reported a confirmed overall response rate of 54% at the recommended dose range, with tumor shrinkage in 88% of treated patients. The median duration of response was about 12 months, with some responses lasting beyond two years. The treatment was generally tolerable: nearly all patients experienced cytokine release syndrome, but most cases were mild to moderate, and there were no treatment-related deaths.²² In a broader analysis across solid tumors including melanoma, a subset of melanoma patients treated at the highest dose levels showed a confirmed response rate of 50%, with several responses still ongoing past 12 months at data cutoff.²³

The bispecific approach works differently. Brenetafusp (also known as IMC-F106C) is an ImmTAC molecule: a bispecific protein with one end that recognizes the PRAME peptide on tumor cells and another end that grabs CD3 on nearby T cells, physically bridging the immune cell to the cancer cell to trigger killing.²⁴ Unlike the T-cell therapy, this is an off-the-shelf drug that does not require manufacturing personalized cells for each patient. In a phase 1 study in cutaneous melanoma patients who had already failed checkpoint immunotherapy, brenetafusp showed a disease control rate of 58% among PRAME-positive patients, with about a third achieving confirmed tumor shrinkage. Crucially, none of the five PRAME-negative patients showed any tumor reduction, and their survival was markedly worse, reinforcing that PRAME expression on the tumor is not just an eligibility marker but a genuine predictor of benefit.²⁵

Both therapies currently require patients to carry the HLA-A*02:01 tissue type, which is the most common HLA-A allele in people of European descent (found in roughly 40-50% of that population) but less common in other ethnic groups. This limits the eligible patient pool and is an active area of research as developers work toward versions that recognize PRAME peptides presented by other HLA types.

Non-Invasive PRAME Testing

Standard PRAME testing requires a tissue biopsy, which involves cutting out a piece of skin, processing it, and staining it. Researchers have explored whether PRAME can be detected non-invasively using adhesive patches applied to the skin surface to collect cellular material, followed by PCR testing. An early concordance study found that PCR on non-invasively collected samples agreed with traditional tissue-based staining in about 88% of lesions that were PRAME-positive on both platforms. Interestingly, PCR picked up PRAME in 45% of lesions where the standard tissue stain was negative, suggesting that the molecular test may detect low-level expression that falls below the visual threshold of immunohistochemistry.²⁶ Whether that additional sensitivity translates into clinically useful information or simply more false alarms is still being worked out, but the possibility of testing suspicious moles without a scalpel is appealing.

Why PRAME Expression Varies Between Tumors

A question that runs through all of this research is why some melanomas express PRAME heavily while others barely express it at all. Part of the answer is epigenetic: the methylation status of the PRAME gene promoter determines whether the gene is accessible for transcription, and tumors that have lost this methylation tend to have higher expression.³ But the age-dependent pattern seen in pediatric versus adult melanomas suggests that additional developmental and biological factors are at work. Tumor subtype matters as well: desmoplastic melanomas, which tend to have different immune microenvironments and mutational profiles, are characteristically PRAME-negative, while metastatic deposits often express PRAME at rates equal to or higher than the primary tumor.⁹

For uveal melanoma, the connection between PRAME and metastatic potential appears tied to broader epigenetic reprogramming. Class 2 uveal melanomas, which carry the worst prognosis, are more likely to be PRAME-positive, and within that group, PRAME positivity identifies the tumors at greatest risk of spreading.²¹ This pattern suggests PRAME expression is not just a passive marker but reflects a biological state of the tumor that favors dissemination. The fact that PRAME did not correlate with BRAF or NRAS driver mutations in cutaneous melanoma further indicates that it operates through a pathway independent of the most commonly targeted oncogenic drivers.¹⁹

PRAME Testing in Practice

If you have had a biopsy of a suspicious mole and your pathology report mentions PRAME, it is worth understanding what the result does and does not tell you. A diffusely positive PRAME stain strongly supports a melanoma diagnosis, especially when other histologic features are ambiguous. A negative stain does not rule melanoma out, particularly if the lesion is on a palm or sole, has spitzoid features, or the patient is young. PRAME is an ancillary marker, meaning it adds weight to the overall assessment rather than delivering a verdict on its own.

For patients with uveal melanoma, PRAME status increasingly factors into surveillance planning. A PRAME-positive result, especially combined with a class 2 gene expression profile, signals a higher risk of metastasis and may justify more frequent liver imaging and consideration for adjuvant clinical trials. For cutaneous melanoma, PRAME’s prognostic value has not been established strongly enough to change management independent of tumor thickness and other standard staging factors.

On the treatment side, the clinical trial data for PRAME-targeted therapies are early-phase but genuinely encouraging, especially for patients whose disease has progressed after standard checkpoint immunotherapy. The confirmed response rates in the 50-54% range in checkpoint-refractory melanoma stand out because this is a population that has limited remaining options. Whether these responses translate into long-term survival gains will require longer follow-up and larger randomized trials, but the combination of reasonable tolerability and durable responses in some patients has made PRAME one of the most closely watched therapeutic targets in melanoma research.

1
PubMed. The human tumor antigen PRAME is a dominant repressor of retinoic acid receptor signaling
2
PubMed Central. The role of the cancer testis antigen PRAME in tumorigenesis and immunotherapy in human cancer
3
PubMed. Tumor antigen PRAME is up-regulated by MZF1 in cooperation with DNA hypomethylation in melanoma cells
4
PubMed. Diagnostic test accuracy meta-analysis of PRAME in distinguishing primary cutaneous melanomas from benign melanocytic lesions
5
PubMed. PRAME immunohistochemistry for melanoma diagnosis: A STARD-compliant diagnostic accuracy study
6
PubMed Central. PRAME Immunohistochemistry in Thin Melanomas Compared to Melanocytic Nevi
7
PubMed. PRAME immunohistochemistry as an adjunct for diagnosis and histological margin assessment in lentigo maligna
8
PubMed Central. PRAME Expression and Melanoma In Situ Margin Assessment
9
PubMed. PRAME is a useful marker for the differential diagnosis of melanocytic tumours and histological mimics
10
PubMed. Preferentially expressed antigen in melanoma immunohistochemistry as an adjunct for differential diagnosis in acral lentiginous melanoma and acral nevi
11
PubMed Central. Diagnostic and Prognostic Values of PRAME Immunohistochemistry in Acral Lentiginous Melanoma
12
PubMed Central. PRAME Expression in Mucosal Melanoma of the Head and Neck Region
13
PubMed Central. PRAME Expression Correlates with Genomic Aberration and Malignant Diagnosis of Spitzoid Melanocytic Neoplasms
14
PubMed. PRAME expression in melanocytic proliferations with intermediate histopathologic or spitzoid features
15
The American Journal of Dermatopathology. Pitfalls of PRAME Immunohistochemistry in a Large Series of Melanocytic and Nonmelanocytic Lesions With Literature Review
16
PubMed. Characterization of PRAME immunohistochemistry reveals lower expression in pediatric melanoma compared to adult melanoma
17
Frontiers in Oncology. The Challenge of Melanocytic Lesions in Pediatric Patients: Clinical-Pathological Findings and the Diagnostic Value of PRAME
18
The American Journal of Dermatopathology. The Utility of PRAME and Ki-67 as Prognostic Markers for Cutaneous Melanoma
19
PubMed. PRAME expression in 137 primary cutaneous melanomas and comparison with 38 related metastases
20
PubMed Central. 15-Gene Expression Profile and PRAME as Integrated Prognostic Test for Uveal Melanoma: First Report of Collaborative Ocular Oncology Group Study No. 2 (COOG2.1)
21
PubMed Central. Epigenetic reprogramming and aberrant expression of PRAME are associated with increased metastatic risk in Class 1 and Class 2 uveal melanomas
22
Journal of Clinical Oncology. Phase 1 clinical update of IMA203, an autologous TCR-T targeting PRAME in patients with PD1 refractory metastatic melanoma
23
Nature Medicine. Autologous T cell therapy for PRAME+ advanced solid tumors in HLA-A*02+ patients: a phase 1 trial
24
Immunotherapy Advances. High-affinity T cell receptor ImmTAC® bispecific efficiently redirects T cells to kill tumor cells expressing the cancer–testis antigen PRAME
25
Journal for ImmunoTherapy of Cancer. Phase 1 safety and efficacy of brenetafusp, a PRAME × CD3 ImmTAC bispecific, in post-checkpoint cutaneous melanoma (CM)
26
SKIN The Journal of Cutaneous Medicine. Concordance of Preferentially Expressed Antigen in Melanoma by Non-Invasively Collected Polymerase Chain Reaction and Immunohistochemistry on Paraffin Embedded Tissue