Mycosis Fungoides: Diagnosis and Work-up of Early Stage Disease

Primary cutaneous T-cell lymphoma is a heterogenous group of non-Hogkins lymphomas, including mycosis fungoides (MF), anaplastic large cell lymphoma, adult T-cell lymphoma/leukemia, subcutaneous panniculitislike T-cell lymphoma, and extranodal natural killer (NK)/T-cell lymphoma, nasal type, each uniquely distinguishable based on clinical presentation, immunohistochemistry, prognosis, and treatment strategies. The classification schema is constantly being revised to accommodate new information as it is gathered on this rare set of diseases; however, it is widely recognized that MF is by far the most prevalent of this group,¹ accounting for approximately 50 percent of primary cutaneous T-cell lymphomas and having an incidence of six or seven per one million person-years.² This incidence has been on the rise over the last 30 years, although it is not clear whether this is due to better diagnostic tools (such as polymerase chain reaction [PCR]- based T-cell gene rearrangement technology), heightened awareness among clinicians, or a true rise secondary to an as yet unidentified environmental factor or infectious agent. As such, it has gained more prominence in the dermatologic differential, but poses difficulties to early diagnosis, management choices, and therapeutic options. Thankfully, new treatment choices are now becoming available to the dermatologist that were unheard of 10 to 20 years ago, such as retinoid therapy with bexarotene, histone deacetylase therapy with vorinostat, and many other agents currently in the drug pipeline. In a series of three articles, I will first discuss how to confidently make the diagnosis of MF earlier in the disease course, followed by providing guidance in choosing which therapy is right for any particular patient, including long term management recommendations. Finally, I will focus on helpful tips to successfully manage specific treatment options that may be delivered by the dermatologist.

MF can be fairly easy to diagnose in the uncommon setting of aggressive or rapidly progressive disease. However, diagnosis may be trickier for the other 90 percent of presentations that follow an indolent course that for many years mimics a benign dermatosis. Common benign dermatoses that MF may mimic include atopic dermatitis, psoriasis, allergic contact dermatitis, asteatotic eczema, eczematous/systemic drug eruptions, and small plaque parapsoriasis. We would like to make the diagnosis as early as possible so as to avoid unnecessary interventions, provide correct prognosis, and determine realistic expectations of therapy. In my referral practice, many consultations are for “rule-in/rule-out MF.”

As a student of the schools of cutaneous lymphoma leaders Peter Heald and Joan Guitart, I have found their formulaic approach useful. I am often surprised that what at first blush seemed would fit so well for MF turns out to be excluded by this formula, and vice-versa; what I surely did not originally expect to be MF is actually confirmed to be the correct diagnosis with this method. I look for two-out-of-three features of the presentation to be consistent or classic for MF, so as to be confident in my diagnosis or exclusion. These three aspects of disease are location, morphology and histology. I then grade these three separate characteristics to be classic, suggestive or atypical for MF based upon the patient's presentation. Of utmost importance, given the rarity of MF, a good dose of healthy skepticism for the diagnosis and actively searching for alternative causes of the rash is warranted. My own experience has been that by far and away the most common misdiagnosis is not MF mimicking psoriasis, but actually a drug eruption mimicking MF.

Location
Classic MF primarily arises in the double-clothed areas of the body: axillae, breasts, groin, and buttocks. Allergic contact dermatitis often spares the axillary vault, so it can be a clue in favor of MF if the dermatosis is active in the vault. Although inverse psoriasis may tend to the axillary vault or groin, it typically follows a different morphology (see below) and tends to spare the convex surfaces of the breast or buttock. Once the dermatosis has progressed, other areas of skin may become involved; always ask for the history of where the rash first started. Nearby areas of photoprotected skin such as the extensor forearms or inner thighs may be suggestive of MF. It would be atypical for MF to begin on the face or neck, unless it is the folliculotropic variant.

Morphology
Classic MF primarily presents with thin, red, scaly, poorly demarcated plaques of at least 5cm in diameter. Over time the morphology tends toward thickening of the plaques to more erythematous prominence, and, of course, tumors that may ulcerate. It can be confusing however, when the initial presentation is actually erythroderma, as this is more often associated with drug reaction, seborrheic dermatitis, atopic dermatits, contact allergy, psoriasis, or HIV infection. Suggestive morphology might include numerous small eczematous plaques with only one (or none) larger than 5cm in size, alopecic well-defined plaques, or papules coalesced into plaques that wax and wane, albeit in a set of fixed locations. Atypical primary morphology would be deeply erythematous (hyperemic), targetoid, nodular, or ulcerated lesions.

Pathology
Early MF can be subtle and usually lacks classic histopathologic signs, therein causing so much trouble for making the early definitive diagnosis. This is why I do not rely solely on pathology to confirm early stage disease. On the flip side, when pathology is classic but the other two criteria of location and morphology are atypical, I doubt the diagnosis of MF and search for other (usually drug-induced) causes of the presentation. Classic histopathology of MF reveals all of the following: lichenoid dermatitis with “Indian file” lining up of atypical lymphocytes at the dermo-epidermal junction, exocytosis of atypical lymphocytes, minimal or no spongiosis, wiry collagen in the papillary dermis, and Pautrier's microabcesses (which are nearly pathognomonic). Having a few of these histologic features present is variably suggestive of MF, depending on which ones and how many are present. It would be atypical to have any significant spongiosis or infiltrating eosinophils or neutrophils in number. A granulomatous infiltrate may rarely be seen in the granulomatous slack skin variant of MF, but this should always be accompanied by the other classic histopathologic findings mentioned above.

Immunohistochemistry classically shows a CD3+, CD4+, CD8-, epidermal infiltrate with CD4:CD8 ratio >4 and relative loss of CD7. Other suggestive immunohistochemistry profiles may be present, such as CD3+/CD4-/CD8+ or CD3+/CD4-/CD8- epidermal infiltrates.

Although PCR based T-cell gene rearrangement (TCR) studies to attempt to identify a clonal T-cell population with a single specific V-beta gene rearrangement have become more available and popular, this technology tends not to be helpful in early MF. Early disease is many times pauci-cellular, making it hard to identify a clone out of the background of polyclonal tumor infiltrating lymphocytes. The test also has poor specificity, as a positive result is found in many benign dermatoses that elicit a clonal T-cell population: atopic dermatitis, allergic contact dermatitis and psoriasis, just to name a few. Hence, a positive TCR is only suggestive of MF, and only in the right clinical setting. Of more use is a negative TCR, as the test is very sensitive. So if biopsy material comes from a well-defined and obviously cellular plaque, but TCR is negative, you can with fairly good assurity (greater than 85 percent) rule out MF.

In order to provide the pathologist with the most tissue to view, I will sometimes opt to use a shave technique, instead of punch, for biopsy of what may be early disease. This technique allows visualization of more dermal-epidermal pathology. Even if you regularly read your own slides, I recommend consulting these samples to a dermatopathologist whom you can call to discuss results and who sees many MF cases in consultation each year.

Work-up
This can also be a confusing topic as the guidelines for work-up as outlined by the National Cancer Center Network (NCCN)3 are somewhat onerous for early stage MF. Many dermatologists may also feel the complete work-up is a daunting task to coordinate. The generally recommended complete work-up includes: complete physical exam with attention to skin and lymph nodes, complete blood count with Sezary smear, peripheral blood flow cytometry, complete metabolic profile, LDH, chest x-ray, neck/chest/ abdomen/pelvis CT or whole body PET/CT, and lymph node biopsy of suspicious nodes. In selected cases bone marrow biopsy may also be recommended.

The guidelines actually leave a bit of leeway for using common sense to determine what is appropriate. I do not typically do chest x-ray, peripheral blood flow cytometry, or CT scans to stage what is clinically evident early stage disease (IA-IIA) unless there are concerning factors on physical exam or review of systems that would warrant this level of work-up. In general, the small odds of finding systemic disease in early stage MF do not warrant this level of limited medical resource use. For stage IIB or above, this is definitely a proper staging work-up, as tumors or erythroderma are more highly associated with greater systemic involvement and the findings may alter expected prognosis and treatment options.

In regards to lymph node biopsy, this is helpful if treatment of clinical stage IIA disease does not elicit shrinking of lymph nodes in six to eight weeks of skin improvement, or skin fails to improve with appropriate therapy. Expect to find dermatopathic (enlarged but not containing lymphoma) lymph nodes if greater than 20 percent body surface area is involved with patch or plaque disease. However, no one would fault you for initially obtaining a lymph node sample in stage IIA disease. But if nodal biopsy is warranted, I have a surgeon perform a whole node biopsy for the hematopathologist to characterize architecture and level of involvement, as sampling error is too high with core biopsy and no architecture can be gleaned from a fine needle aspirate.

Once the work-up is complete and staging determined, I believe that care should continue at the direction of the dermatologist. Our field is best educated to deal with skin directed therapies and monitoring of response to therapy, even if some of those therapies end up being delivered by the local oncologist or radiotherapist. Typically, patients should be followed once every three to four months, if not more often, at the start of therapy to help monitor therapeutic responses, side effects and disease course.

Early stage disease may be monitored less often than this if the patient already has shown a long indolent clinical course. Once stable disease is obtained, follow-up every six months is typical. Even after obtaining complete and durable remissions, I follow patients at least once a year for the rest of their lives. The main point here is that care should not completely transfer out of the dermatologist's hands, as I have seen too many patients treated with inappropriately aggressive and ineffective regimes once the dermatologist's perspective has been lost from these patients' plans of care.

Increasing Confidence
Hopefully these tips will aid the dermatologist to correctly diagnose mycosis fungoides earlier in the disease course and manage their own MF patients with greater confidence. In the next installment I hope to provide insight on when to actively treat and how to choose a therapy that is correct for your patient.

Dr. Mark has no relevent disclosures.

1. Willemze R, Jaffe ES, Burg G, et al., WHO-EORTC classification for cutaneous lymphomas, Blood, 2005;105:3768–85.
2. Criscione VD,Weinstock MA, Incidence of cutaneous T-cell lymphoma in the United States, 1973–2002, Arch Dermatol, 2007; 143:854–9.
3. National Cancer Center Network. NCCN Clinical Practice Guidelines in Oncology: Non-Hodgkin's Lymphomas (V.3.2008). Available at: www.nccn.org/professionals/ physician_gls/PDF/nhl.pdf