Novartis Pharma K.K. said Monday it could find no evidence a former employee manipulated or altered clinical research data for the drugmaker's Diovan blood pressure-lowering drug.

Although Kyoto Prefectural University of Medicine said earlier the data had been manipulated, a third-party investigation has found no evidence of manipulation or alteration, said Novartis Pharma, the Japanese unit of Swiss drug company Novartis AG.

But at a press conference Monday, Novartis Pharma President Yoshiyasu Ninomiya apologized for embarrassing patients, their families, health care providers and people in general with the latest data manipulation scandal.

A former Novartis Pharma employee participated in clinical research on the drug by the university and four others and wrote a clinical research report without specifying the employee's affiliation, leading to doubts about the neutrality of the research.

The investigation carried out by the university found that data had been manipulated for its report concluding that Diovan could reduce cerebral stroke and cardiac angina incidence more effectively than other drugs.

The trials found Diovan, one of the company's top-selling medicines, which racked up sales of $4.4 billion last year, reduced stroke risks in Japanese people.

In February, one of the studies was retracted by the European Heart Journal, which said "critical problems existed with some of the data reported" in the paper. The data could have been manipulated, according to a report by the university on its website on July 11.

The five studies were conducted between 2001 and 2004 at Jikei University in Tokyo; Chiba University; Kyoto Prefectural University of Medicine, and Nagoya University.

The scope of Novartis' probe into the research, and whether it was falsified, is limited because Novartis does not have access to raw data obtained during the studies, Ninomiya said.

The company will work with universities to determine what happened, he said.