EDITORIAL - Malignancy and rheumatic disease - a real association

[Guest guest]

Journal of Rheumatology

October 2005

Editorial

--------------------------------------------------------------------------------

Malignancy and Rheumatic Disease - A Real Association?

SHARON CHAMBERS, MBBS (UWI), MRCP (UK),

Rheumatology Specialist Registrar;

DAVID ISENBERG, MD, FRCP,

Academic Director and

Arthritis Research Campaign Professor of Rheumatology,

Center for Rheumatology, The Middlesex Hospital,

University College London,

Arthur Stanley House,

40-50 Tottenham Street, London UK W1T 4NJ

--------------------------------------------------------------------------------

Links between autoimmune rheumatic diseases (ARD) and malignancy go back 100

years: Stertz in 1916 reported the case of a patient who had inflammatory

muscle disease and a gastric carcinoma1.

During the past 30 years, there have been many attempts to determine more

precisely the link between cancer and ARD. The best established connections

are those between Sjögren's disease and non-Hodgkin's lymphoma (NHL)2; and

adult onset dermatomyositis and a variety of solid tumors, notably ovarian,

pulmonary, and pancreatic tumors3-5. More recently, a weak but definite link

has been reported between systemic lupus erythematosus (SLE) and

non-Hodgkin's lymphoma6.

The explanation of this association between most ARD and cancer remains

elusive. Given that both cancer and ARD are relatively common, the

association may be due to chance alone. The possibility exists, however,

that an ARD developing shortly after a diagnosis of cancer is in fact a

paraneoplastic syndrome. Alternatively, the malignancy might be a

consequence of immunosuppressive therapy often used in the treatment of

patients with ARD6.

Whereas this possible link to therapy is clearly a possibility in some

cases, the recently published analysis of the patients with SLE indicated

that the tumors occurred early, not late, in the course of the disease6.

Further, as Sjögren's patients are rarely treated with immunosuppressives,

no drug link can explain the connection with increased risk.

Several studies have, however, indicated an increased risk of

lymphoproliferative malignancies in patients with rheumatoid arthritis (RA)

that may well be related to the use of cytotoxic agents or prolonged

immunosuppressive therapy7,8. Cases of lymphoma occurring in patients with

RA receiving low-dose methotrexate therapy have been reported. In some

cases, the lymphoproliferative disorders occurred in the context of

associated Epstein-Barr virus infection, and tumor regression was noted to

occur after discontinuation of methotrexate9,10.

Additional studies are needed to define clearly the tumor-inducing role of

immunosuppressive therapy, including the newer anti-tumor necrosis factor-a

(TNF-a) agents, in patients with ARD. The definitive answer to the possible

anti-TNF-a links to cancer should come from the biological register

established by the British Society for Rheumatology. Four thousand patients

with RA each taking one of the 3 approved anti-TNF-a drugs and 4000 patients

with RA not taking any of these drugs are being followed in a longterm

fashion.

As indicated above, it is well recognized that patients with malignancies

may develop autoimmune and rheumatic manifestations. A new methodology

called SEREX (serological analysis of recombinant cDNA expression libraries

of human tumors with autologous serum) is adding significantly to our

knowledge of autoantigens11.

Several bona fide studies employing this technique list the role of

tumor-associated autoantibodies against various tissue-specific antigens,

membrane receptors, and nuclear proteins (including tumor suppression

genes)12. Although still embryonic, such pioneering research may pave the

way for a better understanding of autoimmunity, and facilitate early

detection and treatment of malignancies.

It is of interest that non-Hodgkin's lymphoma (NHL) appears to be the

commonest cancer link in several ARD. The article by Cuttner, et al in this

volume of The Journal adds to our knowledge by reviewing this problem13, but

starting from the endpoint of malignancy and working backwards to identify

those patients who had evidence of an autoimmune disease (AID). The authors

have reviewed patients with known NHL, comparing them with patients who have

other hematological disorders to determine the prevalence of associated AID.

The authors did not limit their analysis to patients with ARD, but also

considered diseases such as inflammatory bowel disease and gluten

enteropathy. The inclusion of uveitis may have been inappropriate, however,

without further clarification of the etiology, as uveitis may be of

non-autoimmune origin. It is also not clear what the relevant autoimmune

tests for inflammatory bowel disease, polymyalgia rheumatica, polyarteritis

nodosa, chronic urticaria, pemphigus, and psoriasis are. It would have been

helpful to mention whether tissue (histological) confirmation of these

diagnoses was used where appropriate in addition to clinical features and

serological markers.

It is not entirely clear whether the onset of AID in patients with NHL was

confirmed by chart review or whether the authors relied solely upon patient

report. This doubt obscures the latency period between the diagnosis of AID

and the development of NHL.

Further details about the control group would have benefited this study. The

authors compared the NHL patients with those who had hematological disorders

as well as " other patients. " However, it is not stated what diseases

constituted the group of " other patients, " 43 in all. They noted that the

NHL autoimmune patients were more likely to have had a prior rheumatologic

disease, but the details were not clarified.

The authors observed an association between the use of immunosuppressive

drugs in the NHL patients versus the controls. It is not clear at what stage

these therapies were used, and whether the drugs such as MTX were used for

treating the rheumatic disease or the malignancy. Such information would

have strengthened any association between the two.

It is interesting to note their finding that the most frequent disorders in

NHL patients were RA (n = 7) and autoimmune hypothyroidism (n = 7). Only 2

of the patients with NHL had Sjögren's and they both had Graves' disease.

Four of the patients with RA received immunosuppressives, which might have

been a contributing factor. It is notable that the patients with autoimmune

thyroid disease (n = 7) constitute the (equal) highest group with NHL. Since

they were almost certainly not treated with immunosuppressive drugs and were

also the groups most likely to have an IgG kappa monoclonal antibody, it is

tempting to speculate that these patients may be at particular risk of

developing neoplasia.

Cuttner and colleagues have succeeded in heightening the awareness of

autoimmune disease and NHL in females (20% of all women with NHL compared to

7% of controls) (p = 0.001). They conclude that for patients of the same sex

and age, patients with AID are 2.6 times more likely to have NHL than

controls.

A dedicated attempt to study the link between cancer and AID is certainly a

step in the right direction. Given the potential benefits of detecting early

malignancies, such studies provide a platform for the development of

age-specific screening guidelines for patients at high risk, and for the

development of optimal management protocols for autoimmune diseases.

http://www.jrheum.com/subscribers/05/10/1866.html

Not an MD

I'll tell you where to go!

Mayo Clinic in Rochester

http://www.mayoclinic.org/rochester

s Hopkins Medicine

http://www.hopkinsmedicine.org