American Society of Clinical Oncology (ASCO) 2012 Annual Meeting: Page 2 of 2

July 19, 2012

High Prevalence of Bisphosphonate-Associated Esophageal Cancer Reported by FDARequires Further Investigation

A review of the US Food and Drug Administration (FDA) Adverse Event Reporting System (AERS) uncovered 128 cases of bisphosphonate-associated esophageal cancer, and issued a significant safety signal with alendronate. The same research team, however, failed to find an increased risk of esophageal cancer with bisphosphonate use in a large academic medical center. The researchers, led by Beatrice Edwards, MD, associate professor of medicine, Department of Orthopedic Surgery, Northwestern University Feinberg School of Medicine, Chicago, IL, presented their findings in a poster session at the ASCO meeting, calling for continued investigation into this adverse drug reaction and for increased vigilance when prescribing oral bisphosphonates.

In 2009, the FDA reported on 23 patients who had developed distal esophageal cancer within 2 years of initiating alendronate therapy. In addition, 31 cases of esophageal cancer were reported in Europe and Japan. Esophagitis has been associated with use of oral bisphosphonates; however, a few population studies have failed to find an association between esophageal cancer and bisphosphonate use, said Edwards.

To explore this potential relationship further, the team assessed the FDA AERS for a safety signal for bisphosphonates and esophageal cancer by searching for terms related to esophageal cancer and drug names for bisphosphonates, using a search period of 1996 to 2010.

Of the 128 patients (average age, 72 years) who were identified from AERS as having developed esophageal cancer following bisphosphonate use, 75% (n=96) of these patients had taken alendronate. A significant safety signal was found only for alendronate, with a proportional reporting ratio of 6.4 (P=.001), corrected for time of exposure (alendronate was the first oral bisphosphonate to gain FDA approval).

In an attempt to verify the association, “we wanted to look at a clinical database,” Edwards said. The team therefore conducted a cohort study within the Northwestern University clinical database, which contains 2.4 million individual participants with information about medication use. The total number of bisphosphonate users in the data warehouse was 15,621. Twenty-six cases with esophageal cancer had at least two prescriptions for a bisphosphonate, but in only six of these cases did the bisphosphonate use precede the esophageal cancer. There were 1384 cases of esophageal cancer reported without bisphosphonate use. The odds ratio for bisphosphonate-associated esophageal cancer was 0.6, which was not elevated or significant.

A case-control analysis of these six cases with 183 controls on bisphosphonates who did not develop esophageal cancer revealed that gastroesophageal reflux disease (GERD), hiatal hernia, gastritis, and a history of peptic ulcer disease were reported more commonly in individuals with esophageal cancer.

“Most of us worry about giving bisphosphonates to a patient with Barrett’s esophagus [a condition in which the esophageal lining is damaged by stomach acid] because it’s premalignant, but we’re thinking that we may also want to be more careful with our GERD patients as well, and not be giving them oral bisphosphonates,” Edwards said. “In patients with a history of peptic ulcer disease or active GERD, I’d probably go with an intravenous bisphosphonate or another medication, but I don’t think it’s cause for great alarm. We do need to continue pharmacovigilance.”—Wayne Kuznar



Early Functional Decline During Chemotherapy Predicts Worse Survival in Older Patients With Cancer

Loss of autonomy early during cancer treatment is linked to poorer prognosis in older patients. A standard oncologic evaluation of older adults may lead to overtreatment and an excess risk of toxicity, or conversely, undertreatment and loss of efficacy. The geriatric evaluation may be “a good solution to solve these problems,” as it provides an opportunity to better evaluate risks, to better appreciate chances, and to propose tailored treatment strategies, said Pierre-Louis Soubeyran, MD, PhD, Institut Bergonie, Bordeaux, France, during a clinical science symposium at the ASCO meeting.

The appropriate treatment objective in the older cancer patient is optimal tumor control while maintaining quality of life, Soubeyran said. Important events to avoid that may have an impact on treatment strategies are early death, early functional decline, and hospitalization for toxicity. In a multicenter prospective study, researchers set out to examine whether a decrease in autonomy for activities of daily living (ADLs) after a first cycle of chemotherapy influences prognosis in older patients with cancer.

The investigators sought to precisely define and identify predictors of early functional decline. “The Activities of Daily Living Scale is definitely the right tool to use,” said Soubeyran. “The question is, when to use it? We decided to evaluate it early; that is, before the second cycle of chemotherapy.” A small threshold of a reduction of ≥0.5 points on the ADLS (scores range from 6 to 0) was used to define early functional decline.

The study enrolled 364 patients aged >70 years (average age, 77.3 years; 59.2% men) with various kinds of cancer (excluding breast cancer) who were receiving first-line chemotherapy. Of these patients, 299 were evaluable.

Prognostic factors were sought from the pretreatment geriatric assessment data, which included the Cumulative Illness Rating Scale-Geriatric (CIRS-G), Instrumental ADL (IADL), Mini Nutritional Assessment (MNA), Mini-Mental State Examination (MMSE), 15-item Geriatric Depression Scale (GDS15), 30-item Quality of Life Questionnaire (QLQ-C30), Eastern Cooperative Oncology Group Performance Scale (ECOG-PS), and Get Up and Go test, as well as from patients’ baseline biological and clinical information (ie, age, sex, tumor extension and localization, performance status, body mass index, weight loss, albumin level, C-reactive protein level, hemoglobin levels, leukocyte and platelet count, and creatinine clearance). Four geriatric assessments were performed: before treatment, before cycles two and four, and at the end of treatment. Patients completely dependent at baseline  (ADL score of 0) were excluded.

Fifty patients (16.7%) experienced early loss of autonomy. At a median follow-up of 5 years, loss of autonomy was associated with a 52% increased risk of death compared with no early functional decline (P=.016). Biological and clinical factors were not associated with loss of autonomy. A low score on the GDS15, the number of dependencies on the IADL, a low score on the MMSE, a slow time on the Get Up and Go, a low score on the ECOG-PS, and a low score on the MNA pretreatment were found to be prognostic of early functional decline in univariate analyses. In the multivariate model, a low score on the GDS15 (odds ratio [OR], 2.4; P=.01) and dependencies on the IADL (OR, 3.0; P=.027) were independently associated with an increased risk of early functional decline.

“Early functional decline has some prognostic value in terms of survival [in elderly patients with cancer], so it’s probably wise to think about using the ADL questionnaire during treatment for these patients,” concluded Soubeyran. “The GDS15 and IADL questionnaires should be evaluated at baseline,” he said, as these scores are predictive of early functional decline.—Wayne Kuznar