By David N. Leff

What can lung cancer learn from prostate cancer?

Ask molecular biologist David Misek, a senior research associate at the University of Michigan, Ann Arbor: ¿Let¿s say you¿re a cigarette smoker,¿ Misek said, ¿and you go to your primary-care physician every year for a checkup. The doctor takes a blood sample, and looks for the presence of antibodies to a pair of proteins called annexins. If antibodies to one or both of them show up, that may warrant a more diagnostic study to determine whether you have an early lung tumor or not. It¿s along the same lines as the prostate-specific antigen (PSA), which assays for prostate cancer.¿

Misek is second author of a paper in today¿s Proceedings of the National Academy of Sciences (PNAS), dated Aug. 14, 2001. Its title: ¿An immune response manifested by the common occurrence of annexins I and II autoantibodies and high circulating levels of IL-6 [interleukin-6] in lung cancer.¿ Its senior author is research oncologist Samir Hanash, a professor on the university¿s faculty.

Annexins are a family of at least 13 calcium-dependent phospholipid-binding proteins They are thought to mediate intracellular calcium signals, and their genes act as tumor suppressors.

¿Identification of circulating tumor antigens or their related autoantibodies,¿ the PNAS paper begins, ¿provides a means for early cancer diagnosis as well as leads for therapy.¿

¿We found that over 60 percent of serum from patients with lung cancer,¿ Misek told BioWorld Today, ¿contained antibodies against annexin I or II, which are a set of cell-surface glycoproteins. But none of the non-cancer controls exhibited IgG-based reactivity against proteins identified as glycosylated annexins I and/or II.

¿Annexin I is a target for autoantibodies in autoimmune diseases, such as systemic lupus erythematosus,¿ he added, ¿whereas annexin II has never been implicated previously as a target for autoantibodies. Annexin II has been involved in cell-cell adhesion as in plasminogen activation. It may function as a cell-surface receptor. The two annexins have different molecular weights, and bind to different receptors.¿

A Messenger of Tumor Inflammation

As for interleukin-6, it is a cytokine derived from macrophages and epithelial cells. It increases synthesis and secretion of immunoglobulins by B lymphocytes. ¿This IL-6 is secreted by infiltrating cells of the immune system,¿ Misek said. ¿In any given tumor tissue, you have a mixed population. Blood cells, white cells, other inflammatory cells come in. IL-6 is part of that inflammatory response. As we said in the PNAS paper, when we treated a lung tumor cell line with IL-6, it actually caused an increase in the amount of annexins I and II in the tumor cells¿ membrane compartment.

¿This whole PNAS finding,¿ he observed, ¿has novel clinical utility as a diagnostic tool for the early detection of lung cancer in patients. Obviously, the earlier the detection, the better off the patient may be.¿

Lung cancer killed an estimated 160,000 Americans in the year 2000, according to the American Cancer Society. This mortality far exceeded the expected 129,400 deaths from breast, colorectal and prostate cancers combined. A patient with an untreated case of lung cancer will die on average in eight months. The five-year survival rate is 13 percent.

To assay the presence of annexin autoantibodies in lung cancers, the PNAS co-authors procured serum and freshly excised tumor tissue from the University surgical department. The 54 patients with adenocarcinoma, squamous cell, small cell and large cell carcinomas included 29 males and 25 females, with an age range of 46 to 82 years. Of this cohort, 16 subjects (30 percent) had annexin I autoantibodies in their blood, and 18 (33 percent) annexin II, for a total of 63 percent. By comparison, 60 other patients, with brain, breast, liver, esophageal cancers and melanoma displayed six annexin I autoantibodies, and zero annexin II. Fifty-one healthy subjects scored zero on both counts. ¿Interestingly,¿ the PNAS paper recounted, ¿IL-6 levels were significantly higher in sera of antibody-positive lung cancer patients compared with antibody-negative patients and controls.¿

In all cohorts, there was no correlation between smoking status and the occurrence of autoantibodies to annexin I and/or II in patients with lung cancer. Among subjects with autoantibodies, 89 percent were smokers, while of those without antibodies, 96 percent were smokers.

¿There is some evidence that occurrence of autoantibodies in lung cancer is of prognostic relevance,¿ Misek said. ¿Just as with the prostate PSA prognosis, the earlier you can detect annexin autoantibodies in lung cancer, the better the prognosis will be. We have not yet done a correlative study as to whether patients with higher levels of antibodies will have a worse prognosis or not. That¿s something we need to do soon. We also have other prognostic studies of lung cancer where we¿re looking at a whole bunch of genes and their expression levels.¿

Clinical Trials? Still A Bit Nebulous

¿The good thing about the prognosis or early detection is that we did not find any correlation with early-grade tumors ¿ stage I tumors, for example, which have a good five-year survival rate ¿ or stage II tumors, which have much poorer survival. We found 51 percent of patients with stage I lung cancer had antibodies to one or another of the annexins; 67 percent with stage II, and 43 percent with stage III. So there did not seem to be a progression of antibodies with the stage of the tumor. Rather, it seemed that if a patient had a tumor early on, he or she developed the autoantibodies against the disease.

¿We are going to be looking more at the prognosis, and how patients with antibodies in the serum do in terms of survivability,¿ Misek said. ¿We also would like perhaps to use some of this information to design clinical trials, although this is still a little bit nebulous as to what to really go after.¿

¿Although there is much interest in the identification of antigens that induce a cytotoxic T-cell response,¿ the PNAS article summed up, ¿the identification of panels of tumor antigens that elicit an antibody response may have utility in cancer screening or diagnosis or in establishing prognosis. Such antigens may also have utility in immunotherapy against the disease.¿