College of American Pathologists
Printable Version

  Feature Story


cap today

Cytokines’ steady march to labs

May 2000
Karen Titus

If cytokines labored under any slogan, it might well be the popular political directive, "Think globally, act locally."

Taking the global perspective is easy. Pick a problem, and cytokines are likely to be involved: Cancer. AIDS. Crohn’s disease. Rheumatoid arthritis. Sepsis. Nosocomial infections. Diabetes. Asthma. Graft-versus-host disease. Hypertension. Colitis. Postpartum anemia. And the list keeps growing.

Despite their far reach, however, cytokines appear to be most successful clinically when they have a limited locus of operation. Moreover, no one mechanism is likely to explain their function and possible use. This is no field for dyed-in-the-wool absolutists. Administering exogenous cytokines may be helpful in some cases—and in other cases can kill. Cytokines can be inhibited and replaced—but in many cases should not be. Their presence may indicate spread of disease—or reveal absolutely nothing.

Confused yet? Hang on—it’s only going to get worse.

"Never make an absolute in cytokine research," says Marion C. Cohen, PhD.


"Never," she repeats firmly.

Even the mere act of defining "cytokine" eludes simplicity. In an interview with CAP TODAY as well as in a presentation given at last year’s American Association for Clinical Chemistry annual meeting, Dr. Cohen explained the ongoing puzzle that is cytokine nomenclature.

Cytokines were originally named for the activity they described, which resulted in a plethora of three-, four-, and five-letter acronyms and perpetual confusion as researchers deciphered cytokines’ complex actions. "It was unclear whether each cytokine activity represented a distinct cytokine or multiple activities of fewer cytokines. As it turns out, it’s some of both," reports Dr. Cohen, associate professor of pathology and laboratory medicine, UMDNJ-New Jersey Medical School, Newark.

Early definitions characterized cytokines as cell-derived, hormone-like polypeptides that regulate cellular replication, differentiation, or activation in processes relating to host defense and repair. The next step in the definition evolution was this pronouncement: Cytokines were cell-derived regulatory polypeptides involved in the coordination of multicellular interactions. This was followed by the view that cytokines were macromolecules that influence cellular behavior.

"Then it became ’a molecule with biologic effects,’ and, finally, ’a molecule,’" says Dr. Cohen, not entirely facetiously.

For sticklers who insist on a definition, the current generally agreed-upon concept describes cytokines as simple polypeptides or glycoproteins with molecular weights of usually less than 30 kd, Dr. Cohen says. For many years cytokines were called hormone-like, although the distinction is blurring, with growth hormones and prolactin considered to be cytokines—at least by some.

Cytokine production is transient, and the action radius is typically short. Cytokines create their actions by binding to specific high-affinity cell surface receptors. Though individual cytokines can display a broad and diverse range of actions, at least some actions of each cytokine are targeted at hemopoietic cells. And while researchers used to believe all cytokines were secreted, "We know today there are at least some membrane-bound forms of some of the secreted cytokines, and at least one chemokine that seems to be completely membrane bound," Dr. Cohen says.

Many—but not all—cytokines have multiple target cells and actions. Redundancy is common with some groups of cytokines; different groups may have similar actions. Synergy as well as antagonism can occur. Simultaneous exposure of cells to multiple cytokines can lead to qualitatively different responses; likewise, a cytokine may increase, decrease, or otherwise affect production of another cytokine or the expression of another cytokine’s receptors.

Classification of cytokines can be somewhat arbitrary as well. In her AACC presentation, Dr. Cohen outlined the following categories:

  • cytokines that mediate and regulate innate immunity, such as type 1 interferons; tumor necrosis factor-alpha; interleukin-1, -6, -10, -12, and -15; and chemokines (although not everyone subscribes to the chemokines-are-cytokines theory).
  • cytokines that in general mediate and regulate specific immunity, including IL-2, -4, -5, -13, -16, and -17; interferon-gamma; transforming growth factor-beta; and lymphotoxin (also known as tumor necrosis factor-beta).
  • cytokines that stimulate hematopoiesis, which include IL-3, -7, -9, and -11; and the colony-stimulating factors.

Cytokines can also be grouped on the basis of their receptors, according to the sequence homology of the receptors to which they bind, Dr. Cohen says. Families of receptor proteins include the immunoglobulin superfamily; cytokine receptor 1 and 2, or the hematopoietins; the TNF receptor family; and the family of receptors involved in chemokines. Not surprisingly, the families are often further divided into subfamilies.

All this might be a matter of interest only to chemistry genealogists if some general rules of behavior could be applied to cytokines. They can’t. Thus, anyone who wants to understand the function and possible clinical utility of a particular cytokine better be prepared to give that cytokine fairly close scrutiny.

"The only thing you can say with any assurance about cytokines is that the old notion we had—that you can identify a specific cytokine as having a specific function physiologically or a specific pathology—doesn’t seem to happen. What you have is a smorgasbord of cytokines and a given reaction that you have to study," says Stanley Cohen, MD, who is credited with minting the word "cytokine" as well as being the first to describe cytokine function (Cohen S, et al. Cell Immunol. 1974;12:150 and Cohen S, et al. Mechanisms of Cell-Mediated Immunity. New York: Wiley; 1974:331-358).

Measuring cytokines presents pathologists with an ample melange of choices as well. "We’ll probably never have a situation where if you have one single value it absolutely says you have this one single disease," says Daniel Remick, MD, professor of pathology, University of Michigan, Ann Arbor. "It’s always going to be a complicated matter, because so many dynamic interactions take place, on so many different levels."

Yet Dr. Remick and others are convinced clinical laboratories will be called on to measure cytokines for their clinical colleagues. Understandably, they just don’t know when. Or how. Or why. At least not yet.

"I see a really, really big role for pathology—at some point," says Dr. Stanley Cohen, who chairs the Department of Pathology at UMDNJ-New Jersey Medical School, Newark.

"We’re going to be asked to do this for a number of reasons," says Dr. Remick, who also spoke about cytokines at the AACC meeting. Cytokines and their inhibitors are slowly but surely creeping into the clinical setting, he notes. At the same time, as researchers further unravel the link between cytokines and various diseases, measuring cytokines undoubtedly will assist clinicians with differential diagnoses.

In fact, Dr. Remick reports, his laboratory has already begun fielding requests from clinical colleagues to test for cytokines. Predictably—given cytokines’ omnipresence—the requisitions come from all over. "We hear from rheumatologists and cancer specialists," Dr. Remick says. "People doing clinical trials. Trauma burn surgeons. Gastroenterologists. You name it."

Reading about cytokine research in the literature seems to prompt many of the requests, Dr. Remick says. "They read about a study in a journal and say, ’Oh, if I could measure this, then I could determine what to do next.’"

Likewise, says Dr. Marion Cohen, the word about cytokines is beginning to leak into the popular press. "A few weeks ago I saw an article about medical treatments in New York magazine, of all places, and it talked about cytokines. I said to my husband [who happens to be Dr. Stanley Cohen], ’Look, cytokines are entering prime time for the public.’

"On the other hand," she’s quick to add, "they’re certainly not ready for prime time for most treatments and certainly not for most tests."

How soon might cytokine testing take center stage? It’s hard to say, experts agree.

"We’re closer than we were," says Monica L-S Tsang, PhD, vice president of research at R&D Systems, which manufactures research reagent kits for measuring cytokines. "But we are still far away. It’s simply a matter of time before measurements of certain cytokines will be useful for certain diagnoses, but we don’t know how long that will be. Maybe years."

Or maybe not. Says Lance Fors, PhD, president and CEO of Third Wave Technologies: "There’s a tremendous amount of interesting information that will be coming out in the literature in the next six months that will help drive the initial utilities of this into the clinical situation." Though he declined to be more specific, given that the studies were pending publication, he noted his predictions were based on his company’s collaborations with researchers propelling the need for cytokine assays.

"We’re out there in the field and are a lot closer to new developments than most people," he says. "So there’s no doubt in my mind cytokines are moving toward the lab, and it’s coming much, much faster than most people realize. There’s some exciting discovery work going on, and it’s pretty late-stage. And it’s going to drive what happens in the clinical labs."

Researchers have unveiled scores of new findings related to cytokines, some, all, or none of which may beat a path to the laboratory and clinical settings. Picking the likely winners requires beating odds that would make most bookies smile greedily.

"It’s hard," admits Dr. Marion Cohen. "Everybody’s got their favorite cytokine."

She gives the nod to chemokines. "There seems to be a lot of effort expended in this area right now," she says, particularly with regard to the role chemokine homologues or chemokine receptor homologues play in certain viral diseases. "We also know HIV uses some chemokine receptors as cofactors for entering cells," she adds, "so this is an important area for HIV research."

Among other favorites, Dr. Stanley Cohen says he’s intrigued by the potential impact of cytokines in embryogenesis differentiation. "This is going to be tremendously interesting," he predicts. "Many of the factors involved in influencing and controlling this are regulated by cytokines, which may make their appearance very early in immunologic development."

A by no means exhaustive (though possibly exhausting) overview of promising cytokine developments includes the following.

Cancer. One current hope is that cytokines will be useful at least as an adjunct therapy for cancer, by modifying them and delivering them in what Dr. Remick calls "industrial-strength doses."

Interleukin-2 and tumor necrosis factor are among the most widely used cytokines for cancer therapy and have met with varying success, Dr. Remick reports. Originally used by itself, IL-2 now is primarily used in conjunction with lymphokine-activated killer cells or tumor-infiltrating lymphocytes. IL-2 has been used extensively in treating patients with renal cell carcinoma, with the first clinical trials—some of which also involved interferon—being done in patients with advanced or widely metastatic disease. Response rates run between 20 and 30 percent, he reports, "which doesn’t sound great until you remember that these were patients who failed everything else." IL-2 has also been used in patients with advanced or widely metastatic melanoma; response rates again range between 20 and 30 percent.

TNF embodies the ignominy of cytokine failure as well as the promise of success. "We originally thought TNF-alpha would possibly be capable of killing tumor cells and under certain conditions would be the treatment of choice," Dr. Marion Cohen says. "But by and large that hasn’t panned out." Ironically, one reason for TNF’s failure has been its own importance, which has made it difficult to pin down for clinical use. "It does so many things, and it has proved to be such a critical cytokine, capable of causing good as well as harm under various circumstances," she says. "That’s one of the biggest problems with cytokines."

The dose-limiting toxicities of TNF have also stunted its application, although it can be used in isolated limb perfusion to avoid amputation in patients with widely metastatic disease. In this setting, surgeons isolate the endangered limb with a tourniquet, cannulate the appropriate artery and vein, and administer high-dose TNF in addition to chemotherapeutic agents. "Clinically the response is spectacular," says Dr. Remick.

Toxicity has plagued other cytokines as well. IL-12 given systematically as a recombinant protein caused severe toxicity (including two reported fatalities) in a phase II clinical trial, for example, which halted further clinical development despite its intense antitumor activities.

Yet researchers have suggested at least one possibility for circumventing that problem. In a recently published study (Narvaiza I, et al. J Immunol. 2000;164:3112-3122), a Spanish team found that adenoviral vectors expressing the chemokine interferon-gamma-inducible protein-10, or IP-10, and suboptimal doses of IL-12 extinguished tumors in a mouse model of colorectal cancer.

"Our results are very encouraging because of the potency of the antitumor activity that we observe even at very low levels of IL-12 gene transfer," says Ignacio Melero, MD, PhD, of the University of Navarra School of Medicine, Pamplona, Spain. "The overall synergistic effects were able to eradicate large established murine tumor nodules."

The team is now preparing for several clinical trials, one of which will use a human version of a recombinant adenovirus encoding for IL-12 genes. "The clinical setting for our studies is very challenging, for there is not current treatment for such malignancies at those advanced stages," Dr. Melero says, adding that U.S. groups plan to use similar gene transfer technology in other tumors, such as melanoma, that traditionally have been more responsive to immunotherapy.

Granulocyte colony-stimulating factor and IL-11 are among other cancer-related cytokine success stories, according to Dr. Remick.

IL-11 reduces the need for platelet transfusions following chemotherapy, in addition to reducing the time needed to recover from thrombocytopenia. G-CSF, which increases and enhances the number of circulating neutrophils, has been indicated for use in chemotherapy, bone marrow transplantation, inducing progenitor cells, and radiation therapy, as well as for treating febrile neutropenia and for some nonneutropenic settings.

Inhibiting cytokines related to angiogenesis is another "hot area of research right now," says Dr. Stanley Cohen.

At the same time, researchers recently have demonstrated that, when elevated in plasma, two angiogenesis-related cytokines—vascular endothelial growth factor and hepatocyte growth factor—may indicate the presence of a tumor.

The investigators tested plasma levels of five cytokines, including VEGF and HGF, in 75 patients with a variety of solid tumors. Statistically significant differences in VEGF and HGF levels were found in subjects who had local disease compared with those who had metastatic disease.

"In the case of VEGF, this is likely to reflect the increase in angiogenesis that accompanies tumor growth," explains Michael S. Pepper, MD, PhD, one of the researchers involved in the study, which was published earlier this year (Fuhrmann-Benzakein E, et al. Int J Cancer. 2000;85:40-45). "In the case of HGF, this is likely to reflect the increase in tumor bulk—many of the tumors were carcinomas, which are epithelial in nature, and in which HGF is likely to be an important tumor cell growth factor." Therefore, he adds, "I would think that measuring both is optimal."

Dr. Pepper, senior lecturer in the Department of Morphology, University of Geneva Medical Center, Switzerland, notes the timing of blood sampling was critical. He and his colleagues observed an augmentation in plasma concentrations in the early postoperative period for VEGF in breast and ovarian carcinomas, and for HGF in ovarian and hepatocellular carcinomas, followed by a decrease when measurements were made on samples collected two weeks later. The early increase, they suggest, may be associated with normal wound healing. The researchers are now setting their sights on a more in-depth prospective study of selected common tumors—lung, colon, breast—using serial measurements.

Congestive heart failure. Compelling evidence exists that inhibiting TNF may be useful in treating congestive heart failure in patients with high plasma levels of TNF. Phase II and phase III trials are underway and appear promising, says Dr. Remick. "It’s too early to say, of course, but it may offer a whole new class of agents for treating this disease."

One caveat, he cautions, is that patients with Crohn’s disease or rheumatoid arthritis, who also respond well to TNF inhibitors, do not have elevated TNF levels. Those diseases may be driven by local concentrations of TNF, rather than plasma levels. "It may be a little bit easier with congestive heart failure, because the heart is pumping the blood, giving you a better chance of having access to the local levels that are causing the congestive heart failure."

Transplantation. Cytokines are associated with disregulations as well as disease. Polymorphisms in the promoter regions of various cytokines may be predictors of clinical response to liver transplantation, for example. "If this turns out to be a common pattern, molecular diagnostic labs will be looking for cytokine polymorphisms," predicts Dr. Stanley Cohen.

Dr. Fors adds that a number of studies suggest a donor’s specific expression pattern or specific polymorphism for a given cytokine could be as valuable as HLA tissue typing for predicting transplant success.

In other work, researchers have shown that an antibody targeted against the gamma-chain common to the five known T cell growth factor receptors—IL-2, -4, -7, -9, and -15—allows stable engraftment of pancreatic islet transplants by inducing apoptosis of activated T cells.

Reporting in a recent issue of The Journal of Immunology (Li XC, et al. 2000;164:1193-1199), Xian Chang Li, MD, PhD, and colleagues hypothesized that blocking the signaling component represses T-cell activation, permitting long-term engraftment.

"Most people have thought that in order for the graft to survive, the only thing you need to do is turn off the immune system with a potent immune suppression," says Dr. Li, assistant professor of medicine at Harvard Medical School, Cambridge, Mass. "This is what we practice in the clinic: We give a patient as many drugs as we can to suppress the immune system, and then we hope the graft will survive."

That may not be the best approach, he suggests. "Immune tolerance is a learned process," he postulates, one that requires new T-cell activation. This requires engagement between the foreign antigen expressed on the graft and the host T cell; once it occurs, only T cells activated by the graft will be allowed to die, creating a very selective tolerance.

In the study done by Dr. Li and colleagues, mice treated with a control antibody rejected the allografts within 17 days; 75 percent of those treated with the common gamma-chain antibody had graft survival beyond 150 days.

Use of the antibody against the gamma-chain has a significant drawback: Its low affinity may be unable to compete with the high-affinity cytokines for binding. Instead, the better approach might be to block the signaling process mediated by the common gamma-chain, using, for example, a JAK3 inhibitor to bypass the competing antibody and cytokines. It’s an approach that has already caught the eye—and presumably dollars—of a number of major pharmaceutical companies, he adds.

Diabetes. Another mouse model has explicated the role of cytokines in the pathogenesis of autoimmune diabetes.

Pere Santamaria, MD, PhD, and colleagues recently published a study in The Journal of Clinical Investigation (2000;105:459-468) that suggests proinflammatory cytokines promote diabetogenesis by marking pancreatic beta cells for Fas-dependent destruction by highly pathogenic, autoreactive CD4-positive T lymphocytes. However, beta cells in the animals normally do not express Fas at sufficiently high levels to be detected by lymphocytes. "So something else had to be happening," says Dr. Santamaria, associate professor, Department of Microbiology and Infectious Diseases, University of Calgary, Alberta.

The missing piece of the puzzle "turned out to be some cytokines," he says, namely interferon-gamma, TNF-alpha, IL-1-alpha, and IL-1-beta. When the cytokines are secreted at the site of inflammation—in this case the pancreas—they stimulate the pancreatic beta cells to express higher levels of Fas on the surface. This permits the lymphocytes to engage Fas and trigger apoptosis. "It’s as if they paint the beta cells for death," Dr. Santamaria says. Evidence from human tissue studies suggests Fas may be implicated in the destruction of human pancreatic beta cells as well, he adds.

Inhibiting the expression of these cytokines early on may delay onset of diabetes, he suggests. "Obviously there are exceptions, but in general terms, interfering with the expression of interferon-gamma or TNF can delay or prevent onset of the disease in animal models."

However, blocking only one cytokine is not sufficient. "If you impede one, the others will still upregulate Fas expression on the beta cell—you have to figure out how to suppress them all to achieve a complete effect. And that’s difficult to do."

Rheumatoid arthritis and Crohn’s disease. Soluble receptors for TNF are considered to be a major therapeutic advance in treating rheumatoid arthritis. In 1998 they were given the green light by the Food and Drug Administration for treating this disease in adults; they’re presently on the FDA fast track to be approved for juvenile rheumatoid arthritis, Dr. Remick reports. The receptors appear to be most effective when given in conjunction with methotrexate, and recipients of this therapy show a significant decrease in the number of swollen and tender joints.

Also in use for treating rheumatoid arthritis—usually in conjunction with other therapies—are monoclonal antibodies that bind to and inactivate TNF. "Patients exhibit a significant decrease in the number of tender joints and in the pain score, and a significant decrease in C-reactive protein," Dr. Remick says.

Anti-TNF is establishing a strong track record in treating Crohn’s disease as well. In one study, patients who failed high-dose steroids, cyclosporine, combination therapy, and other conventional treatments exhibited an 80 percent response rate when treated with anti-TNF, with significant drops in their Crohn’s disease activity index as well as CRP levels.

Anemia. Erythropoietin has been "a real success story" for the treatment of anemia secondary to chronic renal failure, Dr. Remick says. Some 95 percent of patients with chronic renal failure respond to erythropoietin—which boosts the number of red blood cells in the peripheral blood by stimulating progenitor cells in bone marrow—with an increase in their hematocrit. Following two months of treatment, virtually all patients are transfusion independent.

This cytokine also appears to effectively treat anemia secondary to AZT therapy for human immuno-deficiency virus. One study showed it reduced the cumulative number of transfused blood units by 40 percent. And when administered to surgical patients two weeks prior to surgery, erythropoietin may reduce the need for packed red blood transfusions, maintain hematocrit levels, and speed recovery.

Hypertension. High levels of transforming growth factor-beta-1 in African-Americans may explain their higher risk of hypertension and related complications, according to a recently published report. (Suthanthiran M, et al. Proc Natl Acad Sci U S A. 2000;97:3479-3484).

Researchers measured TGF-beta-1 protein levels and TGF-beta-1 mRNA levels in peripheral blood mononuclear cells; they also performed TGF-beta-1 codon 10 genotyping in Caucasian and African-American hypertensive and normal subjects.

TGF-beta-1 protein levels were higher in hypertensive subjects (261 ng/mL) than in normal subjects (188 ng/mL), and highest in African-American hypertensives (322 ng/mL). Among normotensives, TGF-beta-1 protein levels were also higher in blacks (221 ng/mL) than in whites (165 ng/mL). A similar pattern occurred in TGF-beta-1 mRNA levels, although the racial differences were not statistically significant. The researchers hypothesize that TGF-beta-1 along with other growth factors may directly cause high blood pressure and/or endothelial fibrosis and vascular injury.

How might clinical laboratory tests fit into these various scenarios?

Take your pick. It could be as simple as an ELISA, or as complicated as PCR. Or a bit of both.

R&D Systems’ Dr. Tsang, also a speaker at the AACC meeting, outlines several methods for measuring cytokines. At the protein level, approaches include:

  • bioassays, which measure the biologically active molecule.
  • immunoassays, which detect not only the biologically active molecule but the denatured inactive cytokine and cytokine precursors, and fragments.
  • functional immunoassays, which can use the extracellular domain of a cytokine receptor, which in turn binds only to the active cytokine.
  • receptor-binding assays, which are another form of functional assays and which use cell fragments, rather than a living cell, for the binding study.

Immunoassays and functional immunoassays are by far the most specific and sensitive, can be done quickly, are cost-effective, and are amenable to automation and high-throughput applications, Dr. Tsang notes. Because immunoassays measure active and inactive cytokine fragments and precursors, manufacturers often substitute receptor bodies—basically recombinant proteins—for antibodies in both competitive and noncompetitive ELISA formats. These bind to the biologically relevant active form of the cytokine but not to precursors or fragments.

The majority of available cytokine immunoassays and functional immunoassays are primarily for research use. Because many of them are still in development, they’re not widely standardized. "It’s not like getting a glucose, where you know what it means coming from any hospital in the country," says Dr. Remick. Cytokine levels can vary depending on the reagents used, how the assay is performed, the presence of reagents or proteins in human plasma, interference from soluble receptors, complex structure, and so on. That makes it "a little hard to use a value from the literature for making therapeutic decisions, because it’s hard to know what your value is compared to what the literature value is," he adds.

Serum plasma samples are especially problematic because of their many interfering substances. Dr. Pepper points to the significance of measuring plasma values versus serum values in his study of VEGF and HGF in cancer patients, which used R&D’s sandwich ELISAs to assess the cytokine levels.

"Many reports prior to our own used serum for their measurements," he says. "However, platelets contain large quantities of VEGF, and serum values are therefore unlikely to be a true reflection of circulating levels of VEGF." The same does not appear to hold true for HGF, however.

Because cytokines rarely act alone, measuring panels of cytokines on multiplex systems may be the most reasonable approach, says Dr. Tsang. "Yet when you’re trying to measure many cytokines simultaneously, you don’t know whether conditions for one assay may interfere with the other, so we suspect it will work for some combinations but not others."

R&D primarily continues to develop kits for single cytokine measurements, she reports, but some clients have begun requesting panels for multiple, custom cytokines. R&D also manufactures what it calls RNA quantikine kits, which measure levels of cytokine messenger RNA.

In addition to other applications, Third Wave has used its Invader platform to look at cytokine gene expression as well as cytokine polymorphisms, says Dr. Fors. "We’re developing a full panel that will probably encompass up to a hundred cytokines and chemokines for quantitative gene expression analysis." The company currently has a panel of cytokine assays, about 15 of which are available to pharmaceutical partners.

Dr. Melero, of the University of Navarra, foresees any number of possible roles for laboratories should cytokine gene therapy move beyond the research realm.

"Biopsies of tumors will be informative to assess whether an immunological reaction is taking place—for instance, taking a look at the inflammatory infiltrate at the treated solid tumors," he says. "Genetic and immunohistology analysis of tumor cells can disclose if escape variants of the tumors are being selected out. Traditional blood tests will monitor the patients in order to detect adverse effects."

The possibilities for therapeutic drug monitoring are not so clear, Dr. Remick says. "Intuitively it makes sense that you’d want to monitor therapy. On the other hand, usually optimum doses are determined by dose-ranging studies. Having said that, though, there are a number of situations where there may be an optimal window for delivering cytokine inhibitors, so it’s quite possible that testing of this sort may evolve, similar to measuring cyclosporine levels or trough gentamicin levels."

Moreover, testing cytokine levels prior to starting a therapy "is probably worthless, at least right now," he ventures, in part because the correlation between plasma levels and disease activity is not necessarily a strong one, particularly if disease activity is localized, as is the case with rheumatoid arthritis, Crohn’s disease, and diabetes.

Given the multiple variables involved in any cytokine equation, it’s understandable why testing possibilities remain vague. Even in situations where a cytokine level clearly corresponds to a disease state, testing may not be worthwhile. High levels of IL-6, for example, invariably indicate sepsis. "But most clinicians don’t need a blood value to make a decision about whether the patient is septic or not—they can tell by looking at the patient. And they probably wouldn’t treat on the basis of that individual value," says Dr. Remick.

Yet hope for cytokine testing springs eternal. While attending a medical meeting in Munich in March, Dr. Remick says he learned neonatologists in Europe are routinely using plasma IL-8 levels in lieu of neutrophil counts to determine the presence of nosocomial infections in their patients. "It’s exciting, because clinicians are using IL-8 numbers to make a decision about whether to give antibiotics or not."

Perhaps such actions represent a solid step forward for cytokines after multiple false starts. "There’s been a lot of disappointment in cytokine research, but I think we’ve bottomed out and are finally on the upward side of the curve—although I’m sure others may disagree," says Dr. Marion Cohen.

If nothing else, researchers have learned one sure lesson, she says. "The easy things didn’t work. Now we’ve got to work a little harder."

Karen Titus is CAP TODAY contributing editor and co-managing editor.