Aspirin taken regularly after — but not before — the diagnosis of colorectal cancer appears to reduce the likelihood of recurrence and the risk of death from colon cancer, at least for those patients whose cancer overexpressed cyclooxygenase 2 (COX-2).

Over a period of almost 12 years, there were 81 colorectal cancer deaths among 549 patients who started taking aspirin after the cancer was diagnosed compared with 141 deaths among 730 patients who didn’t use aspirin after diagnosis — a colorectal cancer death rate of 15% versus 19%, according to a study published in the Aug. 12 issue of the Journal of the American Medical Association.

The overall mortality rate was also better among patients who initiated aspirin use following diagnosis, 35% versus 39%, wrote Andrew T. Chan, MD, MPH, of Massachusetts General Hospital and colleagues .

• Explain to interested patients that this report describes findings from an observational study, so the results cannot be the basis for clinical treatment recommendations.
• Explain to interested patients that the benefit observed in this study was limited to patients whose tumors over-expressed COX-2, an enzyme that aspirin blocks.

By contrast, patients who used aspirin before diagnosis did not receive a benefit for colorectal cancer-specific death (HR 1.05, 95% CI 0.80 to 1.37) or for overall mortality (HR 0.93, 95% CI 0.77 to 1.11).

After correcting for other factors, patients who added daily aspirin to their postdiagnosis routine reduced the risk of death from colorectal cancer by almost 30% compared with nonaspirin users (HR 0.71, 95% CI 0.53 to 0.95).

The benefit, the researchers said, was driven by patients whose tumors overexpressed cyclooxygenase 2 (COX-2). As a nonselective blocker of cyclooxygenase, aspirin blocks both COX-2 and COX-1.

Chan and his colleagues were able to perform histological analysis on tumors from 459 participants; 314 of those patients had COX-2 positive cancers.

“Among participants with COX-2-positive tumors, regular aspirin use after diagnosis was associated with a lower risk of colorectal cancer-specific (multivariate HR 0.39, 95% CI 0.20 to 0.76) and overall (HR 0.62, 95% CI 0.42 to 0.93) mortality, whereas postdiagnosis aspirin use was not associated with a lower risk of either colorectal cancer-specific or overall mortality for those with COX-2-negative tumors,” they wrote.

The benefit was observed across stage I, II, or III cancers, which “suggests that aspirin may have a specific effect on the prevention or progression of micrometastases among individuals with established disease,” Chan wrote.

Those findings emerged from an analysis of data from a prospective cohort study of 1,279 men and women with stage I, II, or III colorectal cancer who were enrolled in two ongoing, observational studies of health professionals — the Nurses’ Health Study and the Health Professionals Follow-up Study.

And that’s the rub. Although the data are impressive, they are not persuasive because, as observational studies, they do not rise to the level needed to change guidelines, said Daniel G. Haller, MD of the University of Pennsylvania Cancer Center.

Haller, who is editor-in-chief of the Journal of Clinical Oncology, said the findings, added to a series of studies that have demonstrated a chemopreventive effect of aspirin, were compelling so that they make one “think that this is true,” but “we like to have level one evidence.”

Michael J. Thun, MD, MS, vice president emeritus of epidemiology and surveillance research at the American Cancer Society, agreed with Haller. But while lacking conclusive proof that aspirin improved survival, the study advanced the aspirin hypothesis because, Thun said, “the survival benefit associated with aspirin treatment was observed only in people whose tumors overexpressed COX-2 enzyme. This strengthens the indication that aspirin actually caused the benefit.”

In an editorial that accompanied the study, Alfred I. Neugut, MD, PhD, of the Herbert Irving Comprehensive Cancer Center at Columbia University in New York City, wrote that although the findings do require confirmation, “in conjunction with the wealth of data in the precancerous setting, the family findings in the prior study in the Cancer and Leukemia Group B (CALGB trial), and the extraordinarily specific COX-2 biomarker findings bring an observational study as close as it can to offering patients a way to help treat themselves.”

But patients treating themselves is not a good idea, said Michael Hall, MD, director of the gastrointestinal tumor risk assessment program at Fox Chase Cancer Center in Philadelphia.

Asked if it were “prudent” to recommend aspirin as adjuvant therapy for patients newly diagnosed with colorectal cancer, Hall said, “On the contrary. Front-line chemotherapy often lowers platelet counts (thrombocytopenia) making patients more likely to bleed. Adding aspirin, a potent platelet inhibitor, into the mix could set patients up for a serious bleeding episode.”

Chan said that although the study had a number of strengths, most notably “prospectively collected data on aspirin use,” it was limited by its observational design and, although the biological plausibility was “strong,” it was also possible that the observed benefit was related to “the reason for which participants used aspirin.”

The number one reason for aspirin use, Chan said, was for pain relief, but it was possible that “aspirin use may be reflective of other occult predictors for improved prognosis.”

And although the authors had data on causes of mortality, they did not have data on cancer recurrences and they were unable to obtain data on tumor histology for all participants.

The study was supported by the National Cancer Institute, the National Institutes of Health, and the Damon Runyon Cancer Research Foundation. Chan and his colleagues reported no financial disclosures. Neugut reported no financial disclosures. Haller said he had no financial conflicts “regarding aspirin.” Thun reported no financial disclosures. Hall reported no financial disclosures.

This article was developed in collaboration with ABC News.

Primary source: Journal of the American Medical Association

Source reference:

Chan AT et al “Aspirin use and survival after diagnosis of colorectal cancer” JAMA 2009; 302: 649-59.

Additional source: Journal of the American Medical Association

Source reference:

Neugut AI “Aspirin as adjuvant therapy for colorectal cancer a promising new twist for an old drug” JAMA 2009; 302: 688-89.

Earn CME/CE credit
for reading medical news

FDA staff has expressed concerns that denosumab, the investigational biologic drug for osteoporosis, may increase risk of serious infections through its activity against an important immune system modulator.

The agency believes the drug — provisionally trade-named Prolia — could delay fracture healing as well. Some evidence suggests it could also promote tumor development and progression.

The concerns were revealed in briefing documents released in advance of a Thursday meeting of the agency’s Advisory Committee for Reproductive Health Drugs, which will consider whether to recommend denosumab for approval.

Denosumab is a monoclonal antibody to RANKL, the receptor activator of nuclear factor-?B ligand. The molecule appears to help drive osteoclast development and activation, as well as playing a vital role in the body’s defenses against infection.

If NF-?B is the immune system’s “master switch,” as it has been called — controlling B- and T-cell differentiation and dendritic cell development — then RANKL is the finger that flips the switch.

The drug’s manufacturer, Amgen, filed last December for FDA approval of denosumab for preventing and treating osteoporosis in postmenopausal women, in women undergoing hormone ablation for breast cancer, and in men on androgen-deprivation therapy for prostate cancer.

Just this week, two of the registration trials for denosumab were published by the New England Journal of Medicine as early online releases. Those randomized, placebo-controlled studies showed that the drug was as effective as the best bisphosphonate drugs for osteoporosis.

The studies included more than 7,800 postmenopausal women with osteoporosis and close to 1,500 men with prostate cancer.

They also reported no indications of increased rates of serious infections, cancers, or most other serious adverse events, except for an increased incidence of eczema. (See Denosumab a Winner in Phase III Osteoporosis Trials)

The drug also boasts an extremely convenient dosing regimen — subcutaneous injection every six months. Many clinicians believe this is a big advantage for denosumab, considering that compliance with daily oral medications for osteoporosis is notoriously poor.

But the drug’s risk profile, especially with regard to infections and cancers in women being treated for postmenopausal osteoporosis, appears to be the FDA’s chief concern going into the advisory committee meeting.

After pooling data from the clinical studies in postmenopausal osteoporosis submitted by Amgen (including phase I and II trials), agency staff found hints of potential problems.

“Overall, subjects in the denosumab group had a slightly increased incidence of serious infections,” according to the briefing document. “There were more serious infections of the skin, ear, abdominal system and urinary tract. Also, endocarditis, infected arthritis and skin ulcers occurred more commonly in denosumab subjects. There were three denosumab subjects in phase I studies who developed pneumonia requiring hospitalization following a single dose of denosumab.”

The document also noted that while Amgen did not perform carcinogenicity studies in animals because denosumab is not active in normal mice or rats, the clinical data showed a modest increase in certain malignancies in the human subjects.

“Three subjects receiving a high dose of denosumab in [a] dose-finding study died of a new malignancy,” the staff review found. Those individuals received 100-mg doses, whereas 60 mg was used in the phase III studies and would likely be the recommended dose upon approval.

Pooled data from all the postmenopausal osteoporosis trials suggested “a slightly increased incidence” of breast, pancreatic, gastrointestinal, and reproductive-tract tumors.

Twice as many women discontinued denosumab versus placebo because of breast cancer, the reviewers noted — 0.5% (20 cases) of patients receiving denosumab versus 0.3% (10 cases) of the placebo group.

Certain data also indicated that denosumab may produce unhealthy changes in bone structure, the review found.

According to the briefing document, both osteoclasts and osteoblasts were suppressed relative to patients taking placebo and alendronate. Markers of bone dynamics such as activation frequency, bone formation rates, and mineralizing surface were also much lower in denosumab-treated patients.

“This raises a concern that with long term use, suppression of bone remodeling may lead to complications such as delayed fracture healing, osteonecrosis of the jaw, or atypical fracture,” the document said.

Neither of the two New England Journal of Medicine reports this week, however, gave any indication of clinical bone problems associated with denosumab. Authors of those studies said they found no signs of delayed bone healing after fracture and there were no cases of jaw osteonecrosis, a rare but frightening side effect of bisphosphonate drugs.

The FDA document confirmed that no actual cases of jaw osteonecrosis were seen in any of the osteoporosis or hormone ablation trials.

But one case was reported in another trial sponsored by Amgen in patients with multiple myeloma and metastatic cancer.

Also, the clinical data alleviated concerns about possible adverse cardiovascular events, such as promotion of atherosclerosis, that were hypothesized on the basis of its anti-RANKL mechanism.

And, deaths were no more frequent in denosumab groups versus placebo in the trials.

In light of the safety worries that still remain with the drug, the advisory committee will be asked to comment on whether a risk evaluation and mitigation strategy should be required as a condition of denosumab’s approval.

The agency is not bound to follow its advisory committee’s recommendations, but it usually does.

Related Article(s):

• Denosumab a Winner in Phase III Osteoporosis Trials

An investigational biologic drug for osteoporosis increased bone density and reduced fractures in men and women in two placebo-controlled trials.

Three years of treatment with denosumab reduced radiographic spine fractures more than two-thirds in the randomized FREEDOM study of nearly 7,900 postmenopausal women.

Overall, those fractures appeared in 2.3% of the denosumab group, compared with 7.2% of patients taking placebo (P<0.001), according to Steven R. Cummings, MD, of California Pacific Medical Center in San Francisco, and colleagues.

A separate trial by Matthew Smith, MD, PhD, of Massachusetts General Hospital in Boston, and colleagues, documented a similar reduction in new vertebral fractures after two years of denosumab in 1,468 men undergoing androgen-deprivation therapy for prostate cancer, and hence at risk for bone loss.

• Explain to interested patients that standard treatment for osteoporosis, aside from calcium and vitamin D supplements, typically involved bisphosphonate drugs that inhibit osteoclast activity.

• Explain that denosumab also inhibits osteoclast activity but through a different pathway than bisphosphonates and perhaps more completely.

• Explain that these trials were randomized, placebo-controlled studies, the strongest form of medical evidence. Note too, though, that denosumab was not compared directly with a bisphosphonate drug, limiting conclusions about denosumab’s efficacy and safety relative to such existing agents.

Both studies were published online in the New England Journal of Medicine. Both sets of researchers reported increases in bone mineral density (BMD) at various body sites with denosumab, whereas no change or decreases in BMD were seen in the placebo groups.

Preliminary results from the two studies were reported last year. (See ASBMR: Denosumab in Phase III Study Reduces Osteoporotic Fractures and Company Claims Benefit for Drug Aimed at Preventing Bone Loss in Prostate Cancer)

Denosumab is a monoclonal antibody against the receptor activator of nuclear factor-kappaB ligand, or RANKL, a member of the tumor necrosis factor superfamily that is expressed by a variety of lymphoid cells.

This molecule also helps mediate osteoclast development and activation, so researchers believe it tips the balance between bone resorption and bone formation toward the latter, increasing bone density and reducing fracture risk.

The studies “establish the efficacy of a human monoclonal antibody to RANKL, denosumab, in reducing fractures” compared with placebo, wrote Sundeep Khosla, MD, of the Mayo Clinic in Rochester, Minn., in an accompanying New England Journal of Medicineeditorial.

Although the drug did not seem to increase infection rates in either study, Khosla noted that its potential to depress immune function remained an issue.

He added that its cost could “considerably limit its use” if it turns out to be much higher than zoledronic acid (Reclast), the bisphosphonate drug that appears to be denosumab’s most direct competitor.

The double-blind, randomized, placebo-controlled study by Cummings and colleagues enrolled 7,868 women ages 60 to 90 with BMD T scores of -2.5 to -4.0 at the lumbar spine or total hip, reflecting established osteoporosis.

Patients received calcium supplements of at least 1,000 mg/day. Vitamin D supplements were provided as well, with the dose dependent upon 25-hydroxyvitamin D levels. No bisphosphonates or other osteoporosis drugs were allowed.

Denosumab was given as a subcutaneous injection every six months for three years. Lateral spine radiographs were taken annually and analyzed at a central facility for new vertebral fractures.

At the three-year evaluation, the researchers found that the relative risk for new radiographic vertebral fractures in the denosumab group versus those assigned to placebo was 0.32 (95% CI 0.26 to 0.41).

Similar reductions were seen in rates of clinical vertebral fractures (RR 0.31, 95% CI 0.20 to 0.47) and the number of cases involving two or more radiographic vertebral fractures (RR 0.39, 95% CI 0.24 to 0.63).

Other fractures were also reduced significantly, but not as dramatically: the relative risk for hip fracture was 0.60 (95% CI 0.37 TO 0.97) and for nonvertebral fractures it was 0.80 (95% CI 0.67 to 0.95).

The difference between the drug and placebo in vertebral fracture risk was evident during the first year of treatment and was maintained through the succeeding two years, the researchers.

BMD scores in the active-treatment group increased significantly with respect to baseline and to the placebo group, in the lumbar spine as well as total hip.

At the three-year evaluation, denosumab patients had BMD increases of 9.2% in the lumbar spine (95% CI 8.2% to 10.1%) and 6.0% in the total hip (95% CI 5.2% to 6.7%) relative to the placebo group.

In the placebo group, BMD at the lumbar spine barely changed and decreased in the total hip by about 1%.

In the other study, involving men with prostate cancer undergoing antiandrogen therapy, the basic pattern of results was similar, although lumbar spine BMD after two years was the primary endpoint and fracture risks were secondary. As in the FREEDOM trial, the drug was given every six months by subcutaneous injection.

Smith and colleagues found the cumulative incidence of new vertebral fractures in the study after three years of treatment was 1.5% with denosumab versus 3.9% among placebo patients (P=0.006).

Change in BMD at the lumbar spine at year two with denosumab relative to placebo was 6.7% (P<0.001), reflecting a 5.6% absolute increase in the denosumab group versus a 1.0% decrease with placebo.

BMD at other locations, including total hip, femoral neck, and distal radius, also increased from baseline with denosumab and decreased in the placebo group.

Significant differences in BMD were already detectable a month after the first dose, the researchers found. The advantage for denosumab continued to widen through year three of follow-up.

Adverse effects were largely similar between treatment groups in both studies, though the male patients showed slightly higher rates (34.6% versus 30.6% for serious events; 5.9% versus 4.6% for serious events related to infections; P not reported).

In the female study, eczema was significantly more common with denosumab, seen in 3.0% of patients versus 1.7% of those taking placebo (P<0.001).

Notably absent from the list of adverse events was osteonecrosis of the jaw, a rare but worrisome side effect of bisphosphonate drugs.

In an interview, Cummings said the FREEDOM study involved more than 20,000 patient-years of exposure. Estimates of the incidence of jaw osteonecrosis with bisphosphonates range from one in 10,000 to 100,000 patient-years, he said.

Although the data do not rule out the possibility that osteonecrosis could occur with denosumab, it was encouraging to see no cases thus far, he suggested. He said follow-up of FREEDOM participants would continue for 10 years, which may provide a more definitive view of the risk.

In his editorial, Khosla said the efficacy data were similar to those reported for intravenous zoledronic acid and teriparatide (Forteo) in placebo-controlled trials, and better than those for oral bisphosphonates.

But he suggested that compliance could be an advantage for denosumab. “Less than half of patients who are prescribed [current] medications are compliant after one year,” he said, citing a study published earlier this year.

Cummings agreed, pointing out that any nurse or physician can administer the twice-yearly subcutaneous injection of denosumab, whereas zoledronic acid requires an intravenous infusion for which many primary care offices are not equipped.

Teriparatide involves a daily injection that patients can self-administer but may be subject to the same compliance attrition seen with oral drugs.

Khosla pointed to cost as a potential issue for denosumab. Its manufacturer, Amgen, has not announced the pricing. Nevertheless, as a biologic drug, denosumab is widely expected to be relatively expensive.

Khosla noted that zoledronic acid carries a wholesale cost of $1,300 annually for the single annual dose. “Given the relatively marginal clinical differences between these two drugs, a higher cost of denosumab would considerably limit its use,” he wrote.

But another academic specialist contacted by MedPage Today and ABC News suggested a higher price could be acceptable given denosumab’s potential advantages.

Endocrinologist Roberto Pacifici, MD, of Emory University in Atlanta, said the “number needed to treat” in preventing fractures has been low in the denosumab studies reported to date.

“Therefore, the cost of the drug is likely to be [worthwhile] in many patients,” he wrote in an e-mail. “Guidelines will have to be developed in order to use this agent in the most cost-effective way.”

Cumming said that if the drug is approved, he would consider it primarily for patients with a poor history on bisphosphonates.

“Those who have had trouble with oral drugs,” because of side effects or compliance problems, would be the main candidates for denosumab, he said.

Another point in denosumab’s favor, Cummings said, is that it is less persistent in bone than many bisphosphonates, making its activity potentially reversible — an important point for patients experiencing adverse effects.

In addition, Dr. Khosla pointed out, “since bisphosphonates are cleared by the kidney and contraindicated in patients with renal insufficiency, denosumab (which is cleared by nonrenal metabolism) may prove to be a safe drug in these patients, although studies that directly address this issue need to be done.”

Amgen filed last December for FDA approval of the agent for treatment and prevention of postmenopausal osteoporosis in women and treatment and prevention of bone loss in patients undergoing hormone ablation therapy for either prostate or breast cancer.

The two trials were funded by Amgen. Cummings reported relationships with Amgen, Eli Lilly, GlaxoSmithKline , Organon, Pfizer, and Novartis. Other co-authors of the Freedom study reported relationships with a large number of additional corporate entities. Several were Amgen employees. Smith reported relationships with Amgen, GTx, Novartis. Co-authors on that study reported relationships with other corporate entities including Byer, Pfizer, Eli Lilly, Oron, AstraZeneca, Astellas, sanofi-aventis, and Merck. Several were Amgen employees. Khosla and Pacifici reported no potential conflicts of interest.

This article was developed in collaboration with ABC News.

Primary source: New England Journal of Medicine

Source reference:

Smith M, et al “Denosumab in men receiving androgen-deprivation therapy for prostate cancer,” N Engl J Med 2009; 361: DOI: 10.1056/NEJMoa0809003.

Additional source: New England Journal of Medicine

Source reference:

Cumming S, et al “Denosumab for prevention of fractures in postmenopausal women with osteoporosis” N Engl J Med 2009; 361: DOI: 10.1056/NEJMoa0809493.

Additional source: New England Journal of Medicine

Source reference:

Khosla S, “Increasing options for the treatment of osteoporosis” N Engl J Med 2009; DOI: 10.1056/NEJMe0905480.

Earn CME/CE credit
for reading medical news

Related Article(s):

• ASBMR: Denosumab in Phase III Study Reduces Osteoporotic Fractures
• Company Claims Benefit for Drug Aimed at Preventing Bone Loss in Prostate Cancer