Methods of search

Quit Smoking Magic

Quit Smoking Magic

Get Instant Access

The databases of the Cochrane Skin Group, the Cochrane Library to issue 2, 2001, Medline and Embase between 1968 and July 2001 were searched for articles that were trials of therapy of skin disease, or dermatomyositis, or the relationship of dermatomyositis to malignancy. Both Embase and Medline were searched using the Ovid search engine at Nottingham University. The searches involved the following terms:

• the relationship of dermatomyositis to cancer (malignancy, neoplasia)

• treatment of skin disease in patients with dermatomyositis (idiopathic inflammatory myopathy, polymyositis, juvenile dermatomyositis)

• treatment of dermatomyositis with any of the following agents: antimalarials (hydroxychloroquine, chloroquine), corticosteroids (prednisone, methylprednisolone), dapsone, thalidomide, methotrexate, mycophenolate mofetil, azathioprine, intravenous immune globulin (IVIG), probenecid, alendronate, diltiazem, coumadin and sunscreens.

The data are clearer today than they were in 1975 when I first became interested in this issue. However, there are conflicting reports that probably related to the lack of precision in the definition and classification of the patient with dermatomyositis or polymyositis. Population-based studies from Scandinavia clearly demonstrate an increase in the risk of cancer in dermatomyositis, while the modest increase in polymyositis is explained primarily by diagnostic suspicion bias and is not reflected in an increase in mortality (see Table 46.1). It is also clear that these patients are at increased risk for ovarian cancer.6-12 What is not clear is whether these data are applicable to other populations such as Southeast Asians, African-Americans or other ethnic groups.

A recent study from Australia13 calls into question much of the data that has been based on clinical diagnosis. It does appear that dermatomyositis and polymyositis are not the same disease and that their histopathological abnormalities differ significantly and are recognisable by muscle pathologists. Therefore, there may exist a group of patients who were thought to have polymyositis but who might be classified as having dermatomyositis sine dermatitis. This concept is intriguing and might explain why some studies have shown little difference in the prevalence/incidence of malignancy in the two groups. In addition, the existence of this subset might well explain some of the differences that are observed in the studies of therapy (see below).

It is not known whether the increased cancer association is also valid for patients with amyopathic dermatomyositis because the only published data are individual case reports and small case series. One of the difficulties regarding this issue relates to the manner in which ADM is diagnosed. If one applies the criteria of Sontheimer,5 then in my view there are few if any patients with this condition, because either the patient has not been studied in enough detail (magnetic resonance imaging, muscle biopsy, etc.) or the patient has been treated with a corticosteroid and/or systemic corticosteroids.

Several studies have suggested that there might be certain clinical features that are associated


What is the risk of malignancy in the patient with dermatomyositis or amyopathic dermatomyositis (ADM) and how should the patient with dermatomyositis/ADM be assessed for possible cancer?

Study Population- Patients with CA Patients with Statistically Comment(s)

based? In the DM/ADM CA In the PM different?

group group

Study Population- Patients with CA Patients with Statistically Comment(s)

based? In the DM/ADM CA In the PM different?

group group

Slgurgeirsson et al.'*




Both groups had elevated rate i/ controls. Cancer mortality raised only in DM

Ovarian cancer Increased 17-fold

Authors suggest increase In PM due to more extensive evaluation

Alrio et al.7




DM: 6-5 PM: 1-0

Risk rises with age, slight risk for overlap patients Non-melanoma skin cancers, myelofibrosis, polycythemia vera, In situ cervical cancer and 12 preceding cancers eliminated Overrepresentatlon of ovarian cancer

Chow et al.3



reduced to 1-0 by 3rd year following diagnosis

Excess lymphoma/leukaemla

Surveillance for prolonged periods may not be needed

Hill et ai'-




DM: 3-0 PM: 1-3

Plan for evaluation based on the results of their study suggested data derived from Swedish, Finnish and Danish studies, but Included additional follow up from Denmark and Finland

Buchbinder et al.13

Yes, population-based, retrospective cohort study based on pathological findings


Myositis with CTD:


Authors feel diagnosis of DM possible without the presence of a rash, based on the changes observed In the muscle biopsy Increased malignancy rate found In all groups, Including children



Patients with CA in the DM/ADM group

Patients with CA in the PM group

Statistically different?


Stockton et al.20 Yes

Manchul eia/.21 No; CC study

Lakhanpal et ai.22

No; CC study

Lyon et al23 CC study


Women: 30/189 Men: 20/97

11/31 (5 antecedent, 4 concurrent, 2 subsequent)



Zantos etal.24 Meta-analysis of 4 CC 97/513 cohort studies

Marie eia/.25 No; CC study elderly 10/11 elderly v younger patients ?/28 (> 65)


Women: 28/244 Men 12/175

8/40 (3 antecedent, 4 concurrent, 1 subsequent)




1/12-elderly ?/28

P<0-01 (rheumatic controls) P<0-001 (other controls) OR and confidence intervals reported

Not statistically different between groups or i/ controls

No difference i/ controls (siblings)

27 patients with DM and 31 with PM had cancer before DM/PM diagnosed; not known whether any of these patients had metastatic disease at the time of diagnosis

Increase in ovarian, cervical and lung cancer In DM and in Hodgkln's disease in PM found

Controls included cases of skin cancer

Issues about classification for the presence or absence of skin disease

Mayo Clinic studies continue to lack a demonstrable relationship of myositis and cancer

Authors suggest that extensive evaluation Is not needed

Two cases in the control group had skin cancer and should not have been included in the analysis. These patients had a small number of cancers

Increased malignancy In the preceding and subsequent 4 years

The patient with "PM" and malignancy had muscle biopsy changes similar to that expected for DM but lacked a dermatitis

Table 46.1




Patients with CA In the DM/ADM group

Patients with CA In the PM group

Statistically different?


3autas et al,2£i

No; CC study elderly 1/ younger patients (>65)

5/21 1/2/21 Not significant

Chen et al.27


16/91 (0/20 ADM)


P= 0-08

Malignancy was more common In patients over 45;

NPC was the most common cancer (SIR 22-2 1/2-8 for general population)

Malignancy Inversely correlated with Interstitial lung disease

Callen et al.23


7/27 (3 occurred concurrently and 1 within 1 year of DM diagnosis; 3 had prior cancer)

1/31 (Hodgkln's disease preceded polymyositis by 5 years)

P< 0-05

Early study suggesting presence of skin disease associated with cancer

Tymms and Webb23




Not done

Patients with cancer-associated DM/PM were 10-12 years older

Strong association In only 8 patients (4 DM, 4 PM)

Cox et al.30



Not tested

20 patients had an association of cancer and myositis; most discovered by a directed search

High prevalence of ovarian cancer In women and lung cancer In men

Cherin et al."


12/56 (6 ovarian cancers In 45 women)


Not tested

First group to suggest that ovarian cancer Is overrepresented

Table 46.1 (Continued)



Patients with CA In the DM/ADM group

Patients with CA In the PM group

Statistically different?


Davis and Ahmed31

Cross-sectional retrospective review


14 patients with ovarian cancer, 12 with DM and 2 with PM In an analysis of all Mayo cases from 1950 to 1995

Koh et ai.32





Cancer association In older patients, also affected the mortality rate NPC third most common cancer

Maoz et al.33




SIR 12-6 for concurrent and subsequent cancers, but only 4-2 for subsequent cancers alone

3enbassat et al.34




Not done

Prevalence of cancer Increased with age

Wlnkelmann et al.3*




Not done

16/289 patients had cancer

Authors conclude that extensive evaluation Is not necessary

Vloss and Hanelln36




Not done

Focus on negative value of screening by radiological tests

Scaling et a/.10


4/22 (3 ovarian)

1/3 (patient also had myasthenia gravis) cervical in situ cancer

Not done

One of the first reports to note a high prevalence of ovarian cancer Many of these patients had no pelvic examination during Initial hospitalisation

Duncan eta!.37


10/39 had a total of 12 cancers

4 patients with antecedent cancer presented with metastatic disease at the time of DM diagnosis Non-directed malignancy search was not of value Paraneoplastic course was rare

3onnetblanc et al.33


34/118 (7 antecedent, 22 concurrent, 5 subsequent)


Not tested

2 patients with oesophageal cancer, 3 with ovarian occurred concurrently

Table 46.1




Patients with CA In the DM/ADM group

Patients with CA In the PM group

Statistically different?


Peng eia/.3'3





NPC occurred In 12 patients In this Taiwanese study

Whltmore eia/.

4 No

6/19 overall (4/12 with ADM;2/7 with DM)

First linkage of cancer with ADM

Ang et al.t0



ADM 1/ DM Defined as weakness vno weakness, not significant

6/12 had NPC, malignancy occurs In ADM

Defined by clinical or laboratory evidence of myositis, P= 0-0083

ADM, amyopathlc dermatomyositls; CC, case control; DM, dermatomyosltis; CTD, connective tissue disease; IBM, inclusion body myositis, JDM = juvenile dermatomyosltis; NPC, nasopharyngeal carcinoma; OR, odds ratio; PM, polymyositis; SIR, standardised Incidence ratio with a greater risk of malignancy. Fudman and Schnitzer15 reported that patients with dermatomyositis who had a normal creatine kinase had a poor prognosis. Three of their seven patients had malignancy. However, Montagna et al.16 found that 5/17 patients with elevated creatine kinase had malignancy compared with 3/9 with normal creatine kinase. The presence of cutaneous vasculitis may also be associated with an increased risk of malignancy. Feldman et al.17 found clinical evidence of vasculitis manifest as dermal or subcutaneous nodules, periungual infarcts or digital ulceration in seven of 76 patients. Two of these patients had cancer, compared with four of those without vasculitis. In addition, the vasculitic lesions occurred primarily in patients with dermatomyositis (six of seven patients). However, the number of patients in these reports is small and therefore the conclusions reached need affirmation in larger studies. Recently, Hunger etal.18 reported that the prevalence of neoplasia was greater when there was histopathological evidence of vasculitis in the skin biopsy: four of their five patients with vasculitis had cancer, compared with three of their 18 patients without vasculitis (P<0-05). The number of patients in these studies is too small to firmly conclude that vasculitis or normal creatine kinase are markers of malignancy in patients with dermatomyositis.

Another issue that is discussed almost universally in the case-control studies is whether the use of immunosuppressive drugs is associated with an elevated risk of subsequent malignancy. In all of the Scandinavian studies6-9 it appears that there is no increase in the prevalence of malignancy in the patients that have been treated with an immunosuppressive agent; however, there are many individual case reports of subsequent malignancy in dermatomyositis patients. Several reports19 have linked the use of methotrexate with Epstein-Barr virus associated lymphoma. Some but not all of these patients can have a spontaneous resolution of their lymphoma when the drug therapy is stopped.

The search for malignancy in patients with dermatomyositis should include a careful history, physical examination and standard laboratory evaluation (complete blood count, comprehensive metabolic panel, chest radiograph and stool haematest). Any abnormalities found should be thoroughly investigated. In addition, testing should include tests that would be ordered in a "healthy" person of the same age, sex and race as of the patient with newly diagnosed dermatomyositis (for example it is recommended that persons over 50 years of age have a colonoscopy). Data from the recent studies by Hill et al.9 and Stockton etal.20 suggest that for a Caucasian patient with dermatomyositis CT scan of the chest and abdomen, and stool haematest should be performed. In women, pelvic CT and mammography are justified. Tests that are recommended for any person of the patient's age should also be performed (for example colonoscopy in patients over 50 years of age). For patients with polymyositis a chest radiograph and urinalysis should be performed at the time of diagnosis. It appears that continued surveillance is necessary for patients with dermatomyositis, but perhaps not for those with polymyositis. However, what testing should be done beyond age-specific cancer screening is not clear, and the clinician must form a plan in the absence with supporting data. Lastly, nasopharyngeal cancer is much more common in the Asian patient within southeast Asia, and therefore a careful ENT evaluation is needed. It is not known if Asian patients who live elsewhere also harbour this increased risk of nasopharyngeal cancer.

Are there effective treatments for dermatomyositis?

The following are some quotes from "major" dermatologic texts that deal with treatment of patients with dermatomyositis:

• "Most forms of polymyositis and dermatomyositis are approached in a similar manner. Corticosteroids are the main pillar of drug therapy. ... However, early use of corticosteroid-sparing drugs should be considered, including combined therapy, from onset. The main drugs for combined therapy are methotrexate, azathioprine and antimalarials."41

• "Systemic glucocorticoids remain the traditional first line therapy for classic dermatomyositis."42

• "Treatment with corticosteroids is required in almost all cases, the dose depending upon the degree of activity."43

• "Prednisone is the therapeutic mainstay."44

All of the above authorities recommend corticosteroids as first-line therapy, and relatively high doses are generally suggested. In addition, the early use of a corticosteroid-sparing agent is often proposed, with methotrexate or azathioprine being the most frequently suggested agents. Do we know that these medications are effective, what the proper dosing should be, when a second-line agent should be introduced, which agent should be utilised and what the likelihood of success is? Unfortunately, as is illustrated in Table 46.2, the evidence available to answer these questions is poor. It is unclear whether the rate of remission is affected by corticosteroids, whether they are used in high or low doses. In addition, there is little evidence regarding these therapeutic manoeuvres for the treatment of cutaneous disease that might accompany dermatomyositis. Well-controlled randomised trials for this disease are lacking, and even the ones that have been conducted often lack power, or have design flaws or potential biases. Many of the studies have included patients with malignancy-associated myositis, a condition that is believed to respond less well than dermatomyositis. In addition, most studies mix dermatomyositis and polymyositis patients in their analysis, and it has become evident that these disorders are probably different in their pathogenesis and most likely have a differential response to therapy.

Was this article helpful?

0 0
Squash Smoking

Squash Smoking

Squash Smoking Hypnotherapy For Breaking Bad Habits With Smoking Series Presenting your hypnotherapy series for success, enlightenment and breaking bad habits with smoking.

Get My Free Ebook

Post a comment