Lixivaptan for Hyponatremia—The Numbers Game

Abstract

ON SEPTEMBER 13, 2012, THE CARDIOVASCULAR and Renal Drugs Advisory Committee of the US Food and Drug Administration (FDA) reviewed lixivaptan, a new vasopressin receptor antagonist. These agents correct hyponatremia by producing a selective water diuresis without affecting sodium and potassium excretion; the resultant urinary free water loss raises serum sodium levels. Hyponatremia, diagnosed as a laboratory serum sodium of less than 138 mEq/L, is seen in nearly 38% of hospitalized patients and creates a potentially large population eligible for therapy. Historically, severe reductions in serum sodium levels have been accepted as a surrogate end point by the FDA, leading to approval in the United States of both tolvaptan and conivaptan. The label for these approved drugs specifies an indication for the treatment of severe hyponatremia— a serum sodium level of 125 mEq/L—or less marked hyponatremia that is symptomatic and resistant to correction with fluid restriction in patients with syndrome of inappropriate secretion of antidiuretic hormone (SIADH), acute decompensated heart failure, or cirrhosis. However, the validity and clinical utility of serum sodium levels as a surrogate marker for clinical benefit remain uncertain. Lixivaptan demonstrated statistically significant improvements in serum sodium levels at day 7, the primary efficacy end point in populations that had either SIADH or acute decompensated heart failure. The magnitude of the improvement, however, was modest: 1.2 mEq/L in acute decompensated heart failure and 2.4 mEq/L in SIADH. In addition, there were no corresponding improvements in symptoms of hyponatremia as assessed by the Trail Making Test Part B, a measure of cognitive processing speed. Moreover, no improvement in the secondary outcomes of days alive out of hospital or hospital length of stay were demonstrated. In the FDA analysis of the population with acute decompensated heart failure, there was a statistically significant, paradoxical improvement in cognitive processing speed favoring placebo and no observed symptomatic benefit of lixivaptan among patients with SIADH. Of greatest concern, an early imbalance in mortality (15 vs 4 sudden deaths in the first 10 days) was observed with lixivaptan in the population with acute decompensated heart failure, although the randomized study included only 652 patients. Ultimately, the advisory committee voted 5 to 3 against approving lixivaptan for treating patients with hyponatremia who had SIADH, and the committee voted unanimously against its approval for patients with acute decompensated heart failure. The tolvaptan study that led to approval demonstrated a mean increase in serum sodium levels of 5 mEq/L vs placebo at day 4, which persisted at day 30. Tolvaptan also improved the Mental Component Summary of the 12-Item Short Form Health Survey (SF-12) at the 30-day followup. However, this symptomatic benefit was not demonstrated from a validated, disease-specific questionnaire. Moreover, when tolvaptan was tested in a trial of 4133 patients hospitalized with acute decompensated heart failure, no clinical improvements in rehospitalization, renal function, or mortality were observed despite a median follow-up of 10 months. Perceived short-term benefits of tolvaptan improving serum sodium therefore need to be weighed against patients who experienced overcorrection of hyponatremia. The risk of drug-induced hypernatremia was higher in the tolvaptan group than in the placebo group (1.7% vs 0.5%). Although the absolute risk of hypernatremia was small within the randomized trial setting, broader use would most likely increase exposure to this risk of irreversible neurologic injury from osmotic demyelination syndrome without a commensurate improvement in clinical outcomes. Overall, in the case of the vasopressin receptor antagonist class, uncertainty remains regarding both expected symptomatic benefits and long-term improvements in health outcomes. Four explanations were highlighted at the advisory committee meeting that could account for the lack of clinical benefit in the vasopressin receptor antagonist class. First, the degree of hyponatremia may not have been severe enough. Mean serum sodium levels among patients treated with tolvaptan were 128 to 129 mEq/L and among patients in the lixivaptan trials were 124 to 132 mEq/L, yet patients with severe hyponatremia may be most likely to derive benefit. Second, symptoms of hyponatremia may be so