Background and Introduction

Primary cutaneous lymphoma (PCL) is a cutaneous non-Hodgkin’s lymphoma (NHL) that originates in the skin and lacks extracutaneous spread upon initial diagnosis. PCLs do not include lymphomas that secondarily spread to the skin. PCL has a wide variety of subtypes that differ in their clinical presentation prognosis. Indolent-behaving PCLs are slow-growing and less likely to metastasize; therefore, treatment is less invasive and involves primarily monitoring. High-grade PCLs, on the other hand, are more rapid in growth and dissemination and require more aggressive interventions. Conventional imaging techniques such as radiography and ultrasound may be used to screen for clinically indolent PCL. It is recommended that individuals with more aggressive-appearing PCL on physical exam undergo 18F- fluorodeoxyglucose positron emission tomography (18F-FDG PET) combined with computed tomography (CT) imaging to define extent of disease[1]. Treatment options and recommendations depend on both the stage and type of PCL; therefore, it is very important to accurately stage this disease by synthesizing key histologic and diagnostic imaging findings.

18F-FDG PET/CT is an imaging technology that is more specific, as well as sensitive, for detecting both cutaneous and extracutaneous PCL lesions compared to other imaging modalities used in clinical practice. Recent changes in World Health Organization–European Organization for Research and Treatment of Cancer (WHO-EORTC) protocol for staging and monitoring PCL have highlighted 18F-FDG PET/CT’s value as an imaging modality for staging and monitoring aggressive disease due to its unrivaled sensitivity and specificity. This review aims to explore the impact of 18F-FDG PET/CT in classifying, diagnosing, staging, and monitoring PCL.

Incidence and classification

Lymphomas are subclassified into two broad categories: Hodgkin’s lymphoma (HL) and NHL. As of 2018, the WHO-EORTC consensus classification has been used as the gold standard for diagnosis and classification of primary cutaneous lymphomas [2]. PCL is the second most prevalent extra nodal NHL, occurring in 1 out of every 100,000 individuals in the United States. PCLs are classified depending on their specific cell lineage, i.e., T-cell or B-cell origin. Primary cutaneous T-cell lymphomas (PC-TCL) comprise roughly 75% of all PCLs, and primary cutaneous B-cell lymphomas (PC-BCL) comprise the remaining 25% (Fig. 1). Furthermore, PC-TCL and PC-BCL are each subdivided according to clinical manifestation, histopathology, biomolecular profile, and prognosis (Table 1). Fluorescence-activated cell sorting (FACS) is used to determine specific surface antigens present in cutaneous lymphoma cells and further classify these cells by their clonality.

Fig. 1
figure 1

Subclassifications of primary cutaneous lymphomas. PCLs are classified by T-cell or B-cell lineage and subdivided according to clinical presentation, histopathology bimolecular profile and prognosis. PC-TCLs comprise roughly 75% of all PCLs, and PC-BCLs comprise the remaining 25%. Information courtesy of [2]

Full size image
Table 1 PCLs are broadly categorized into T-cell and B-cell subtypes. PC-TCLs generally display more aggressive behavior with poorer 5-year survival rates. PC-BCLs tend to be indolent, slow-growing and have more favorable 5-year survival rates
Full size table

Mycosis fungoides (MF) is the most prevalent PCL, accounting for 60% of all PC-TCLs, and 50% of all PCLs. Variants of MF are distinguished by their unique clinical presentations and histopathologic samples. Some of these variants include pagetoid reticulosis, folliculotropic MF (FMF), and granulomatous slack skin [2,3,4].

Sézary Syndrome is a rare PC-TCL subtype in which identification of peripheral blood involvement is necessary to differentiate it from other PC-TCL subtypes. Primary cutaneous cluster of differentiation (CD) 30( +) T-cell lymphoproliferative disorders (CD30( +) TCLPD) are the second most commonly occurring cutaneous T-cell lymphoma subtype as they contain 25% of all PC-TCLs. CD30( +) variants include lymphomatoid papulosis (LyP) and primary cutaneous anaplastic large-cell lymphoma (PC-ALCL). PC-TCL subtypes other than MF, Sézary syndrome, and CD30( +) TCLPD comprise less than 10% of PC-TCLs (Table 1). The latter include subcutaneous panniculitis-like T-cell lymphoma (SPTCL) and primary cutaneous extranodal NK/T-cell lymphoma, nasal type (PC-ENK/T-NT) [2].

Primary cutaneous B-cell lymphomas are divided into 4 subtypes: primary cutaneous follicular center lymphoma (PC-FCL), primary cutaneous diffuse large B-cell lymphoma leg type (PC-DLBCL-leg type), primary cutaneous marginal zone lymphoma (PC-MZL), and Epstein–Barr positive mucocutaneous ulcers (EBV MCU) (Table 1, Fig. 1). EBV MCU lymphomas comprise less than 1% of all PC-BCLs [1, 2].

Diagnostic methods and staging

The initial evaluation of all PCLs involves a physical examination with skin biopsy and blood draw for smear analysis. A clinician will look for presence of cutaneous lesions, such as erythroderma, plaques, and induration during the physical exam. PCL subtypes can be characterized by lesion location (e.g., trunk, extremities, face, etc.) and the presence of ulceraction. B-symptoms such as night sweats, fever greater than 38˚ Celsius/100.4˚ Fahrenheit for at least one week, and a minimum 10% weight loss within the past six months is suggestive of extracutaneous spread of disease [5]. The lymph nodes (e.g., cervical, inguinal, axillary), liver and spleen, are palpated during the physical exam to determine presence of swelling, as these are common sites to which PCL initially metastasizes. Enlarged lymph nodes or unusual cutaneous lesions should be biopsied. The WHO-EORTC recommends a bone marrow biopsy for patients with intermediate or aggressive subtypes of PCL. 18F-FDG PET/CT imaging should be performed for any patient with suspected extracutaneous disease, or with a predominantly subcutaneous-presenting PCL (Fig. 2) [6].

Fig. 2
figure 2

Schematic diagram of 18F-FDG PET in assessing cutaneous lesions of primary cutaneous lymphoma. Malignant T-cells exhibit increased metabolic activity and greater glucose uptake, which can be detected using 18F-FDG PET

Full size image

Diagnostic blood tests for PCLs include a complete blood count with differential, comprehensive metabolic panel including lactate dehydrogenase (LDH) levels, electrolytes, liver enzymes, and creatinine. Both elevated erythrocyte sedimentation rate and c-reactive protein serum levels are associated with poor prognosis in PCL patients [5]. Patients with erythroderma require peripheral blood taken for smear analysis to determine possible presence of characteristic Sézary cells [1, 2].

Genetic and biomolecular testing to identify cellular markers using FACS is performed to determine the CD4:CD8 cell ratio and PCL clonal subtype (CD3 + /CD7, CD3 + /CD26). Specific CD markers are useful in identifying whether the cell originates from the B-cell or T-cell lineage. This information is used for “blood staging” to determine total tumor burden of the blood [7].

PCL staging uses the tumor–node–metastasis–blood (TNMB) classification system [7, 8] which considers features of disease regarding skin lesions, lymph nodes, visceral organ involvement, and blood tumor burden. A revised TNMB classification guides the clinical staging of MF and Sézary syndrome (Table 2A, 2B) [9], while a separate TNM system is used to classify and stage non-MF or non-Sézary syndrome primary cutaneous lymphomas (Table 2C) [7, 10]. The “T” category describes the severity of skin lesions and is subdivided into T1 through T4. The “N” category describes lymph node involvement, and “M” describes metastasis and if the disease has spread to regional lymph nodes or extracutaneous organs. “B” describes what the “burden” or concentration of these cells is in the blood [1].

Table 2 Clinical staging of MF/Sézary primary cutaneous lymphomas is performed using the TNMB classification (2A, 2B). The clinical presentation of non-MF/Sézary PCLs is significantly different from that of MF/Sézary PCLs, therefore the TNM classification, instead of the TNMB, is used to aid clinical staging of non-MF/Sézary PCLs (2C)
Full size table

TNMB staging of MF and Sézary syndrome primary cutaneous lymphomas

MF and Sézary syndrome must be staged and treated differently than other PCLs because of their unique presentations and progressions. Disease response in the skin is assessed using mSWAT score [11]. A baseline CT is recommended for initial diagnosis of all MF and Sézary syndrome lymphomas, but subsequent scans are generally not recommended in patients with early disease unless nodal spread is suspected. If nodal or visceral spread of disease is suspected, baseline CT should be performed, along with interim imaging to assess early response to treatment, and imaging at the end of treatment to determine success. Stage 1 of MF or Sézary syndrome lymphoma is limited to the skin as patches or plaques. Substages can be classified depending on presence of B-symptoms, lymph node adenopathy, lesion characteristics and peripheral blood involvement.

TNM staging of non-MF/Sézary syndrome primary cutaneous lymphomas

Non-MF and non-Sézary syndrome PCLs are staged using different criteria than those used to stage MF or Sézary syndrome PCLs. The International Society for Cutaneous Lymphomas (ISL) and EORTC has proposed a TNM system of staging that excludes using B-symptoms as a criterion (Table 2C) [10]. This system has a subclassification system using “a”, “b” and “c” to denote varying lesion sizes. “T1” indicates a solitary lesion and is subsequently broken down into two categories: “T1a” lesion < 5 cm in size, “T1b” > 5 cm in size. T2 indicates multiple lesions affecting one or two contiguous body regions and T3 indicates the disease affects the skin diffusely. “N” refers to nodal involvement, with N0 stage being used to classify cancers that lack nodal involvement. “M” indicates whether extracutaneous disease is present.

For all PCL subtypes, the imaging modality chosen for staging is determined by the staging score; T1 skin involvement with B0 blood burden suggests that only a chest radiograph is needed to scan for visceral organ disease. CT with contrast imaging of the chest, pelvis and abdomen is recommended for all higher TNM/TNMB disease scores. As of 2018, the European Society for Medical Oncology (ESMO) strongly recommends 18F-FDG PET/CT imaging for aggressive cutaneous lymphomas, late-stage lymphomas, or lymphomas with signs of extracutaneous disease [1]. TNM/TNMB classifications are only used for staging disease and are not valid predictors of patient prognosis.

Managing indolent versus aggressive PCL

After diagnosis of PCL, subsequent steps in management of disease vary based on whether a cutaneous lymphoma is classified as “indolent” or “aggressive” (Table 1) [2]. Indolent lymphomas in the early stage may be imaged with conventional methods such as a chest X-ray or ultrasound of the abdomen and superficial lymph nodes [12]. CT scan is the recommended imaging modality to assess higher stage PCL lesions which are not 18F-FDG-avid. Whole-body 18F-FDG-PET/CT is strongly recommended for more aggressive or late-stage cutaneous lymphomas such as PC-DLBCL-leg type, PC-FCL, and Sézary syndrome [12]. Staging of disease should be performed annually at minimum to monitor progress of disease. Clinical stage of tumor at time of diagnosis is the most predictive factor when determining disease prognosis. A widely agreed upon, standardized protocol for staging PCL would increase clarity of communication amongst clinicians and more effective collaborative efforts may enable more accurate staging and treatment of PCLs [13]. For PCL, the effectiveness of interventions is highly dependent on appropriate staging of disease; therefore, improving the accuracy of staging with a standardized protocol would, in theory, lead to more effective application of interventions and improved disease outcomes.

The prognosis of all NHLs is predicted using the International Prognostic Index (IPI), which accounts for patient age, LDH levels, location of lesions, and extent of skin involvement [13,14,15]. Subtypes of PCL have individualized criteria to generate a prognostic score specific for that subtype of disease [12]. The IPI score, used to determine treatment in aggressive primary cutaneous NHL subtypes such as PC-DLBCL-leg type and primary cutaneous peripheral T-cell lymphoma, considers updated Ann Arbor criteria, disease stage, serum LDH, hemoglobin, patient age, and patient “performance status” [15,16,17].

Methods

Our search determined that nine new studies investigating the role of 18F-FDG PET/CT in PCLs have been published between 2015 and 2021 (Table 3) [6, 12, 18,19,20,21,22,23,24]. These studies were screened and selected utilizing Google scholar search engine. An advanced Google Scholar search for articles published between the years 2015 and 2021 that contained the exact phrase “primary cutaneous lymphoma” anywhere in the article yielded 885 results. These 885 results contained review papers, case studies and clinical trials used for the results and discussion section of this review paper.

Table 3 Studies reviewed reflect the major updates to the field of 18F-FDG PET/CT in PCL since 2014
Full size table

A nested search was performed to summarize recent clinical trial research focused on defining the role of 18F-FDG PET/CT in diagnosing, staging, and treating primary cutaneous lymphomas. This nested search for articles used advanced Google Scholar search engine to discover all articles published between 2015 and 2021 that contain the exact phrase “primary cutaneous lymphoma” AND at least one of the words “positron emission tomography” OR “PET CT” OR "positron emission tomography computed tomography" anywhere in their text. This search yielded 231 results. Of these results, only scientific studies composed of > nine human subjects, that described 18F-FDG PET/CT detection of cutaneous/subcutaneous PCL lesions, were included. After performing the literature search using the inclusion criteria as described, a total of nine articles remained and were included in this review.

Results and discussion

Other imaging modalities: compared to 18F-FDG PET/CT

Currently, CT is the primary imaging modality used to detect PCL lesions. Although CT is highly sensitive for extracutaneous disease, it has shortcomings in scenarios which require detection of cutaneous malignant lesions. While contrast-CT may delineate lesions, it does not provide functional, metabolic information. This lack of functional data may lead the clinician to overlook small, yet aggressive lesions, subcutaneously presenting lesions such as SPTCL (Fig. 3 and Fig. 4) and PC-ENK/T-NT, and diseased lymph nodes that appear to be ‘normal’ in size [6, 18]. Lack of functional metabolic information is also why CT is prone to false-positive findings in enlarged but benign lymph nodes [25]. Some studies propose that 18F-FDG PET/CT is the preferred imaging modality for diagnosing and staging of PCLs due to its superior sensitivity to CT (Fig. 5) [18, 20, 21, 23].

Fig. 3
figure 3

Subcutaneous SPTCL lesions are easy to miss on CT and visualization is significantly improved with 18F-FDG PET/CT. A 24-year-old female with subcutaneous panniculitis-like T-cell lymphoma located in the upper arms (black arrows). 18F-FDG PET/CT 3-dimensional maximum intensity projection image (A), axial CT (B, E) and axial fusion PET/CT (D, G) images reveal multiple subcutaneous lesions displaying high 18F-FDG uptake in the skin. Images courtesy of Jiang 2021 [6]

Full size image
Fig. 4
figure 4

A 30-year-old man diagnosed with subcutaneous panniculitis-like T-cell lymphoma. 18F-FDG PET/CT three-dimensional maximum intensity projection image (A) and axial fused PET/CT (C, E) reveals multiple focal lesions displaying intense 18F-FDG uptake in the skin. Axial contrast-enhanced CT (B, D) failed to identify these lesions. Images courtesy of Dan 2015 [18]

Full size image
Fig. 5
figure 5

A patient with cutaneous marginal zone lymphoma, an indolent B-cell PCL. Subcutaneous lesions (white arrows) on axial CT (B, D, F) are visible but difficult to detect. These same lesions are more distinct and easier to identify on axial 18F-FDG PET/CT (A, C, E). This study provides evidence that 18F-FDG PET/CT may be superior to CT at detecting not only aggressive PCLs, but also indolent PCLs with subcutaneous presentation. Images courtesy of Davidson 2020 [19]

Full size image

In clinical practice, 18F-FDG PET/CT is not routinely used for disease staging in patients with PCL. Other imaging modalities such as radiography, ultrasound, CT, and magnetic resonance imaging (MRI) are used in combination with physicians’ physical assessment findings to fully stage PCL. 18F-FDG PET/CT is not sensitive for cutaneous lesions with low metabolic activity, which is commonly a characteristic of early-stage PCL and indolent subtypes of PCL [12, 19, 24, 26]. While 18F-FDG PET/CT has not been shown to effectively detect early-stage cutaneous PCL lesions, it is, however, very sensitive for highly metabolically active cutaneous and visceral PCL lesions in aggressively presenting late-stage disease (Fig. 2) [1]. Weighing the cost–benefit ratio of exposing patients to radiation plays an important part in assessment and treatment plan, as clinicians prefer to minimize radiation exposure in patients with limited or indolent disease. As result, until recently the role of 18F-FDG PET/CT imaging in PCL has been limited to patients with aggressive, extra nodal, or primarily subcutaneous disease [1]. However, the latest generations of whole-body PET/CT instruments have a sensitivity 15–68-fold higher than that of conventional PET/CT instruments, requiring lower doses of radiotracer. With whole-body PET imaging, the patient is exposed to far less radiation per image acquired, and the improved safety of this imaging modality broadens its potential clinical applications [27].

To the authors’ knowledge, current methods of diagnosis have not investigated the impact that evaluation of nonattenuation-corrected (NAC) PET images may have on the sensitivity of 18F-FDG PET/CT for detecting cutaneous PCL lesions. NAC PET images are often disregarded; however, NAC PET images can provide invaluable information about superficial lesions and deserve more attention [22, 28,29,30]. As demonstrated in Fig. 6, NAC PET elucidates thinner or less metabolically active cutaneous lesions that are often missed in analysis using solely attenuation-corrected (AC) PET scans (Fig. 6). This is because of the low signal intensity of superficial lesions, which is often lost during the process of attenuation correction of images. It is recommended that nuclear medicine radiologists use the information provided by NAC PET, along with the information provided by AC 18F-FDG PET/CT images, as this would ensure that all cutaneous PCL lesions are accounted for [22, 28,29,30].

Fig. 6
figure 6

3D maximum intensity projection image of an adult diagnosed with PCL. Nonattenuation-corrected 18F-FDG PET images elucidate thin or indolent cutaneous PCL lesions that are often missed on attenuation-corrected 18F-FDG PET analysis. In (A), the attenuation-corrected 18F-FDG PET scan allows for faint visualization of a superficial cutaneous PCL lesion (red arrow) on the anterior left calf. Attenuation correction dampens 18F-FDG signal from superficial cutaneous lesions. In (B), the corresponding nonattenuation-corrected 18F-FDG PET scan of this patient displays the 18F-FDG uptake of this lesion (red arrow) more accurately, allowing for better characterization of its location and metabolic activity

Full size image

The role of 18F-FDG PET/CT in diagnosis and staging

Stage of disease is the most important predictor of disease prognosis in PCL, and thus tumor staging is heavily used to guide the clinician’s generation of an individualized, patient-centered treatment plan. 18F-FDG PET/CT is sensitive and specific for aggressive PCLs, facilitating detection of extracutaneous disease. While 18F-FDG PET/CT influences PCL staging due to its sensitivity for disease manifestation in skin, lymph nodes and extracutaneous tissues, this imaging modality has no value in the pure classification of PCLs. PCLs are classified based on whether they are T-cell or B-cell in origin, and a skin biopsy is necessary for definitive classification (Table 1, Fig. 1).

In centers that have access to this state-of-the-art technology, 18F-FDG PET/CT is quickly becoming the preferred imaging modality for staging PCLs [31]. In 2018, the European Society for Medical Oncology (ESMO) published updated guidelines for the diagnosis and treatment of PCL [1]. Quantification of 18F-FDG uptake in malignant cells may allow clinicians to generate a global disease score to be used for monitoring PCL progression [32]. While 18F-FDG PET/CT has proven itself useful for the monitoring of disease progression in patients with PCL, significant debate persists regarding which imaging modality is best for predicting disease prognosis and monitoring PCL response to treatment. 18F-FDG PET/CT is proven to be very sensitive for extracutaneous spread of PCL, but it has been criticized for its poor sensitivity for indolent cutaneous PCL lesions [1, 21, 24]. However, in subtypes of PCL that display high tendency for metastasis, such as MF or PC-ALCL, it is argued that PET imaging should always be performed at initial staging to determine full extent of disease [21,22,23].

Several recent studies state that utilization of 18F-FDG PET/CT over CT alone would improve the accuracy of initial PCL staging at diagnosis. A study concluded that 18F-FDG PET/CT may increase accuracy of initial staging of PC-MZL, a B-cell PCL that is often missed on physical exam due to its tendency to present primarily subcutaneously. In this study, 18F-FDG PET/CT displayed superior sensitivity for detection of subcutaneous lesions as compared to CT alone (Fig. 5) [19]. A study by Liu et al. determined that 18F-FDG PET/CT is superior to CT or MRI in the initial staging of PC-ENK/T-NT. 18F-FDG PET/CT initial disease staging was more consistent with final staging (confirmed by biopsy) in 94.9% (37/39) of patients, while CT/MRI staging was consistent with final staging in only 74.4% (29/39) of cases [20]. Another study found that in patients with PC-ALCL, initial staging by 18F-FDG PET/CT was more accurate than staging by CT. The sensitivity of 18F-FDG PET/CT for cutaneous lesions was 64% as compared to 18% sensitivity for CT alone[21]. Moreover, a study of 18 patients with non-MF/SS PCL found that 18F-FDG PET/CT has higher sensitivity than CT alone for detection of primary skin lesions, and that CT alone is likely to miss subcutaneous PCL lesions such as SPTCL and PC-ENK/T-NT (Fig. 4). In a cohort of 18 patients, 3 cases of subcutaneous PCL were missed by CT alone; 100% of cases were identified by 18F-FDG PET/CT [18]. Overall, multiple recent studies state that initial staging of PCL is more accurate when performed via 18F-FDG PET/CT than by CT alone. This is true for both indolent and aggressive B-cell and T-cell PCLs.

Guidance of biopsy

Biopsy of the skin and sentinel lymph nodes is performed during the initial evaluation of disease and histological analysis is considered as the gold standard for diagnosing PCL. Currently, the TNM/TNMB classification is used to help guide clinicians determine if, and how, a lymph node should be biopsied (Table 2A, B, C) [7, 9]. Lymph nodes larger than 1.5 cm tend to be classified as ‘suspicious’ and warrant further excision biopsy and analysis [7]. 18F-FDG PET/CT is highly sensitive for metastatic malignancy due to the fact that cancer cells metabolize glucose at faster rates than normal cells and 18F-FDG tracer is radiolabeled glucose. Cancer cells uptake greater quantities of 18F-FDG radiotracer than normal cells, yielding higher-intensity signals on PET images that allow physicians to distinguish cancerous from noncancerous tissue (Figs. 2, 7A, B). Therefore, 18F-FDG PET/CT is utilized for guiding biopsy of lymph node and bone marrow in patients with highly aggressive, late-stage PCLs [1, 22].

Fig. 7
figure 7

Fig. 7A. A 24-year-old male diagnosed with mycosis fungoides. AC 18F-FDG PET/CT three-dimensional maximum intensity projection image (A) displays multiple 18F-FDG-avid, hypermetabolic cutaneous lesions on the anterior chest (B, D), anterior abdomen (C), thighs and face (E). Examination of nonattenuation-corrected 18F-FDG PET/CT images in addition to AC PET images allowed detection of superficial cutaneous lesions not readily visible on CT, with the potential to impact disease staging and patient management. Images courtesy of Alanteri 2015 [22].  Fig. 7B. A 28-year-old female diagnosed with mycosis fungoides. Axial 18F-FDG PET/CT reveals increased 18F-FDG uptake bilaterally in the skin of the lower extremities (A). Arrows indicate bilateral axial (B) and inguinal (C) lymph nodes display intense uptake of 18F-FDG relative to surrounding tissues, reflecting hypermetabolic activity. Images courtesy of Alanteri 2015 [22]

Full size image

Currently, CT is the standard modality used to guide skin biopsy of PCLs. However, recent studies are advocating for more routine use of 18F-FDG PET/CT to guide skin biopsy in both indolent and aggressive PCL subtypes, due to its increased sensitivity for subcutaneous lesions as compared to CT alone. A study investigated the role of 18F-FDG PET/CT in detecting PC-ENK/T-NT, a highly aggressive PCL with mean survival < 12 months. They found that in the detection of malignant PC-ENK/T-NT skin lesions, 18F-FDG PET/CT is 96% sensitive and 98.6% specific, while CT/MRI are 68% sensitive and 97.9% specific. This study concluded that 18F-FDG PET/CT was significantly more sensitive for cutaneous PC-ENK/T-NT lesions than CT alone. They theorize that the routine use of 18F-FDG PET/CT for guiding skin biopsy in this subtype of PCL would accelerate definitive diagnosis, allowing for timely initiation of treatment in a disease with rapid progression and high mortality [20]. Studies by Jiang et al. and Davidson et al. support the novel idea that 18F-FDG PET/CT should be utilized over CT alone for guiding skin biopsy of subcutaneous-presenting, indolent PCLs [6, 19]. Jiang et al. found that SPTCL, an indolent PC-TCL that presents primarily subcutaneously, is highly 18F-FDG avid, making 18F-FDG PET/CT very helpful for guiding biopsy [6]. Davidson 2020 states that while 18F-FDG avidity is variable in indolent PC-MZL, 18F-FDG PET/CT is still more sensitive for the subcutaneous lesions than CT alone [19].

Treatment response evaluation

The Deauville 5-point scale represents the standardized criteria used to interpret and evaluate the treatment response in lymphoma patients [14]. This scale allows the clinician to incorporate semi-quantitative analysis of 18F-FDG PET/CT scans into the assessment of PCL response to treatment [33, 34]. The patient’s mediastinal blood pool is used as a reference for background 18F-FDG tracer uptake, while the patient’s liver is used as a reference for “high” 18F-FDG uptake. Lesions with no 18F-FDG uptake are given a score of 1. Lesions with 18F-FDG uptake equal to mediastinal blood pool are given a score of 2, lesions with uptake less than the liver but more than the mediastinum are given a score of 3, lesions with 18F-FDG uptake moderately higher than the liver are scored 4 and those lesions with significantly higher uptake than the liver are scored as 5. The physician then uses these PET-based scores to label the lesion response to treatment as one of four categories: complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD). CR lesions have a Deauville score of 3 or less and display no bone marrow metastasis. PR lesions display reduced 18F-FDG uptake relative to previous scans and display no structural progression. SD lesions portray consistently elevated 18F-FDG uptake but lack disease progression. PD lesions have a Deauville score of 4–5 and display high 18F-FDG uptake or involve a new 18F-FDG avid focus [34].

The Deauville 5-point scale was designed primarily for use in extracutaneous lymphomas, so it was adopted in combination with the modified severity weighted assessment tool (mSWAT) to be applicable for assessment of PCL scans. Scores for PCL disease progression and response to treatment may be generated using the methods such as mSWAT, Deauville’s score, and 18F-FDG PET/CT. mSWAT is a qualitative method that uses visual assessment to estimate the percent surface area of the patient’s body that is diseased [11]. mSWAT score is weighted × 1 for patch, × 2 for plaque, and × 4 for tumor lesions. The mSWAT score is used to evaluate the response of the patient’s skin to treatment, with complete response indicating 100% clearance of lesions, partial response indicating 50–99% clearance and no new tumors, stable disease indicating < 25% or < 50% clearance of skin disease from baseline with no new tumors, and progressive disease indicating lesions with > 25% mSWAT increase in skin disease from baseline, or new tumors [8]. While technically there is no mSWAT equivalent for non-MF / Sézary syndrome PC-TCLs or PC-BCLs, the mSWAT system may be used [7, 11].

Four of the nine studies reviewed in this paper investigate the utility of 18F-FDG PET/CT in monitoring PCL response to treatment. These studies all used a combination of the Deauville 5-point score and mSWAT to assess disease response [35]. A study by Alanteri et al. showed that 18F-FDG PET allows for easy visualization of disease response to treatment in cutaneous MF lesions (Fig. 8). Review of NAC PET scans improved sensitivity of 18F-FDG PET/CT in cutaneous MF lesions [22]. A retrospective, single-institution study by Davidson et al. concluded that 18F-FDG PET/CT was useful not only in initial staging but also for monitoring subcutaneous MZL lesion response to chemotherapy [19]. Subcutaneous lesions that were difficult to identify on CT appeared prominently on 18F-FDG PET/CT (Fig. 5). 18F-FDG PET/CT detection of hidden subcutaneous MZL lesions at post-treatment follow-up visits changed the disease stage in 8 patients (40%) and resulted in treatment upgrade from radiotherapy to chemotherapy in 2 patients (10%) [19]. A retrospective, single-institution study by Jiang et al. found 18F-FDG PET/CT to be valuable for monitoring SPTCL lesion response to treatment. Due to the subcutaneous nature of this PCL subtype and the superior sensitivity of 18F-FDG PET/CT for detecting subcutaneous lesions as compared to CT alone, Jiang advocates that use of 18F-FDG PET/CT may improve accuracy of disease burden assessment after treatment (Fig. 3) [6]. A retrospective multicenter analysis by Olszewska et al. found that while more than half of indolent PC-BCL (PC-MZL and PC-FCL) lesions are 18F-FDG -avid, the utility of 18F-FDG PET/CT in monitoring disease response to treatment is limited by the variability in 18F-FDG uptake in these lesions [12]. Overall, recent studies conclude that despite the variability in 18F-FDG uptake by indolent PC-BCL cutaneous lesions, 18F-FDG PET/CT continues to stage PC-BCL and PC-TCL with more accuracy than CT alone. The use of 18F-FDG PET/CT at post-treatment visits provides more accurate estimations of PCL disease burden, guiding providers to choose more appropriate treatments and improving the likelihood of positive patient outcomes.

Fig. 8
figure 8

A 50-year-old male with mycosis fungoides. (a) 18F-FDG PET shows 18F-FDG-avid cutaneous lesions at right scalp and right external ear. (b) After treatment with chemotherapy, follow-up PET scan shows complete metabolic response at right scalp and right ear. Images courtesy of Alanteri 2015 [22]

Full size image

One of the most recent scientific endeavors in this field is the effort to define a valid, standardized methodology by which 18F-FDG PET/CT may be used to calculate a semi-quantitative score of global disease in patients afflicted with PCL. 18F-FDG PET/CT is far superior to other imaging modalities in its sensitivity for detecting aggressive cutaneous lesions, suggesting great potential in the clinic. Several studies have attempted to delineate a methodology for using 18F-FDG PET/CT to generate a global disease score [32, 36,37,38] A singular score generated at every follow-up visit may simplify assessment of disease progression, as well as predict the prognosis of disease in patients with PCL. Although a global disease score is a potentially useful clinical parameter, the method is limited by its lack of technical standardization and external validation.

Parameters to track disease recurrence

The recommendation for follow-up frequency may vary and is dependent on the subtype of PCL and its clinical course. Patients with stable disease or indolent PCL are recommended to follow-up with their clinical providers once every 6–12 months, as this allows for screening for disease recurrence. Patients presenting with rapidly progressive or invasive disease may follow-up for disease staging every 4–6 weeks. In addition to repeating the physical exam, peripheral blood is drawn and tested for indicators of disease such as LDH and complete blood count. Furthermore, FACS analysis of CD markers on blood cells may be used to detect disease recurrence. In the case of Sézary syndrome, a peripheral blood smear is analyzed for the presence of Sézary cells. It is recommended that patients with a history of aggressive PCL or extracutaneous disease have 18F-FDG PET/CT imaging performed at follow-up exams to screen for lesions not visible on physical exam [1].

Quantification of primary cutaneous lymphoma/disease activity: global disease assessment

Studies have suggested that the generation of a global disease score representative of patient total body disease burden would be most efficient for tracking progression, recurrence, and therapeutic response of lymphomas [31, 32, 39, 40]. A singular global disease score would be calculated at the patient’s first clinical visit. This score would be re-calculated at subsequent follow-up visits; a reduced global disease score at subsequent visits would correlate directly with a reduction in the patient’s total disease burden, while an elevated score would indicate increased burden of disease (progressive disease/no-response to treatment). Having a singular score representative of patient disease would facilitate the clinician’s ability to determine the course of a patient’s disease accurately and rapidly, and then utilize the most appropriate interventions. Early and accurate staging is extremely important in patients with PCL because treatment type and effectiveness varies greatly based on the stage of disease. Global disease assessments have been shown to provide prognostic information in several other cancers [41,42,43]. Creating objective standards for global disease assessment among institutions can help standardize patient care and augment collaborative efforts. The methodology has previously been applied in clinical studies [44, 45], but despite its obvious advantage it has not been widely adopted in clinical practice until now.

A semi-quantitative approach to 18F-FDG PET/CT would permit generation of a global disease score that is highly accurate and reflective of minute changes in patient disease progression. This score would be applicable for both staging and monitoring of disease progression throughout treatment. Standardized uptake value (SUV) is a semi-quantitative measure representing the concentration of 18F-FDG within the volume of interest (VOI), normalized to the injected radioactivity per unit body weight, and corrected for physical decay [46]. SUV is the most used parameter to quantify metabolic activity in 18F-FDG PET/CT, and maximum SUV (SUVmax) has been frequently used as a quantitative parameter. Initial studies in patients with Hodgkin’s lymphoma showed that changes in SUVmax were more accurate predictors of patient outcomes than changes in the Deauville score [31, 33]. However, because most lymphoma lesions are heterogeneous, the SUVmax is not an ideal measure of tumor metabolic activity as it refers to the voxel with highest uptake within a region of interest (ROI), and hence does not necessarily give an accurate value representative of the disease activity within a region or volume of interest. Several studies have suggested using the mean SUV (SUVmean) instead of the SUVmax to better represent a lesion’s metabolic activity because SUVmean accounts for ROI heterogeneity. Additionally, it was noted that SUVmean values were more susceptible to the partial volume effect (PVE), wherein the limited spatial resolution of PET causes blurring of three-dimensional images and underestimation of 18F-FDG tracer uptake [47]. Lesions that are moving during imaging (for example those on the heart or lung), or lesions that are smaller than the reconstructed spatial resolution (< 1.5–2 cm) are most significantly impacted by PVE, with 18F-FDG uptake being significantly underestimated [47]. Therefore, it is important to perform partial volume correction (PVC) on SUVmean measurements. Metabolic-based volumetric parameters such as metabolic tumor volume (MTV) may be obtained using a threshold to delineate lesion activity [48]. Multiplying the MTV by SUVmean of each lesion provides the total lesion glycolysis (TLG) score for each lesion. Semi-automated software enables the quantification of the SUVmax, pvcSUVmean and MTV for each ROI measured. The MTV or TLG for all lesions in the body may be summed to generate a singular global disease score representative of the total disease burden within the patient’s body.

It is important to note that a criticism of 18F-FDG-PET/CT is its limited ability to detect superficial lesions. However, recent studies propose that analysis of NAC 18F-FDG PET/CT images may allow for identification of cutaneous lesions otherwise not accounted for by analysis of AC PET images alone [22, 28, 30, 49,50,51,52]. Analysis of NAC 18F-FDG PET images alongside AC PET may be necessary to ensure the generation of a global disease score representative of both visceral and cutaneous aspects of total body disease burden (Fig. 6) [22, 28, 30].

Total body PET

Until recently, PET imaging could only be performed on one region of the body at a time. The introduction of whole-body PET systems within the last 2 years has revolutionized clinical approach to staging and monitoring systemic and diffuse diseases. Whole-body PET instruments are not only able to image the entire body in one setting but also up to 68 times more sensitive than conventional PET/CT scanners. The high sensitivity of the whole-body PET allows for decreased dose utilization of radioactive tracer and reduces overall radiation exposure to the patient [47]. Furthermore, it has been proposed that the ideal time point at which 18F-FDG PET/CT imaging should be performed is 2–5 h post-injection [27, 53,54,55,56,57]. This is due to the fact that 18F-FDG continues to accumulate in cells over time and reaches a plateau after 3–5 h; during this time, background tracer also clears, creating an opportune moment to capture a clear image of 18F-FDG uptake in malignant tissues [58]. Due to the increased sensitivity of whole-body PET instruments, a longer duration between tracer administration and image capture does not reduce image quality as it might in conventional PET instruments; whole-body PET can detect 18F-tracers up to 12 h post-injection. This ability to acquire delayed images is important as it optimizes detection of all lesions, including those only detectable after the conventional 1-h uptake time.

Conclusion

Studies reviewed in this paper (Table 3) conclude that the unparalleled sensitivity of 18F-FDG PET/CT for nodal and visceral malignancy makes it a vital tool to accurately determine disease extent, burden, and activity in patients with PCL. The high sensitivity of 18F-FDG tracer for metabolically active skin lesions suggests that 18F-FDG PET/CT may prove useful in guiding skin biopsy in these patients. These studies also conclude that AC 18F-FDG PET/CT is equivalent, if not superior, to CT in detecting cutaneous and subcutaneous malignancy. However, limitations of these studies make this conclusion incomplete. These studies performed analysis of the AC PET images; however, in all but one study, their methodology appeared to lack review of NAC PET images for cutaneous lesions that become undiscernible after attenuation correction. Review of the NAC PET images may significantly improve the sensitivity of 18F-FDG PET/CT for cutaneous malignancy, which can help further obtain a more accurate global disease score calculation.

Primary cutaneous lymphomas are rare malignancies that often go undiagnosed until they have progressed to later stages. Accurate staging of these malignancies early in the disease process is crucial to improving disease prognosis. The diagnosis, staging, treatment, and monitoring of disease in these patients is complex, requiring a multidisciplinary team composed of dermatologists, pathologists, radiation oncologists, nuclear medicine radiologists and hematologists. A standardized method for calculating a global disease score in these patients has yet to be agreed upon; such a score would facilitate accurate discernment of disease progression and allow clinicians to choose more appropriate personalized, patient-centered interventions.