By Frances Bishopp

Hybridon Inc. reported that partner F. Hoffman-La Roche will not continue its antisense collaboration initiated in 1992 to identify lead compounds in both hepatitis C and human papillomavirus (HPV).

Roche, of Basel, Switzerland, has indicated it will assign agreement-related patent rights to the HPV and hepatitis C programs to Hybridon and all licenses granted to Roche under the agreement will be returned to Hybridon. As of March 31, 1997, Roche ceased making research payments to Hybridon, of Cambridge, Mass. The termination of the five-year collaboration, in which Roche purchased a 5 percent equity stake in Hybridon, is effective Feb. 28, 1998.

In late July, Hybridon reported it had stopped further development of its lead product, GEM91, based on data from a Phase II clinical trial of patients with advanced HIV infection. GEM91 (gene expression modulator) is a first-generation (phosphorothioate) antisense oligonucleotide that targets the gag (group-specific antigen) site in the HIV genome.

Hybridon's stock (NASDAQ:HYBN) closed Thursday at $2.875, down $0.125.

Roche had not been able to develop an animal model for the compound for hepatitis C, E. Andrews Grinstead III, chairman and CEO of Hybridon, told BioWorld Today. "They [Roche] thought that was important, but, on the other hand, some people might have looked at this compound and taken it into man right away, even in the Roche organization," Grinstead said.

The work was finished on hepatitis C and Hybridon received the milestone payment, Grinstead continued. The compound for hepatitis C has been shown to inhibit hepatitis C expression in cell culture assays by an antisense mechanism in a dose-dependent manner.

Hybridon intends to pursue development and partnering of both compounds and has identified an animal model for hepatitis C which could be appropriate for in vivo studies.

An animal model was developed for the compound for papillomavirus, Grinstead said, which was the first animal model to successfully demonstrate utility against papillomavirus. "We extensively tested a number of compounds and had active compounds. Those compounds were shown to work over 90 days and over 30 days in the animal model. We had been spending the year trying to perfect a topical formulation," Grinstead said, who added it was becoming apparent that Roche would have continued working on the compounds in such a way that they would not have been candidates until 1998.

"We are disappointed that another Roche program is taking priority here, but we are also very pleased that Roche is willing to give us all the rights to both compounds," Grinstead said. "It was the high road for them to give us the rights back and allow us to move with partners who might have a greater need for these compounds."

Although neither product alone generated much inherent value due to their early stages of development, the termination of this agreement brings about certain risks, Tony Butler, an analyst with Lehman Brothers Inc., of New York, told BioWorld Today.

With no further interest in Hybridon or its projects, at some point Roche might opt to dispose of its shares into the open market, Butler said. Also, those in the investing public with a "glass half-empty" outlook may view this decision by Roche as a lack of confidence in antisense as a platform technology, Butler continued.

"While there may be numerous reasons within Roche itself as to why the project was not renewed, it is not necessarily a discredidation of the technology itself. We continue to believe that second-generation antisense compounds have the opportunity to validate the technology, but this will not occur overnight," Butler said.

Hybridon also is developing GEM132 against CMV (cytomegalovirus) retinitis, which is in Phase I/II clinical trials. This second-generation compound, Grinstead said, has the potential to be dosed very infrequently. "The data we have seen thus far is very encouraging," Grinstead said. Also in development is GEM132 systemic for CMV retinitis, which is in Phase I/II clinical trials.

Hybridon anticipates GEM92 for AIDS, which will be the first potential oral use of antisense, will be in the clinic later this month, Grinstead said. Another second-generation compound, GEM231 (target is protein kinase A) for cancer, will enter clinical trials in the fourth quarter of 1997.

Antisense molecules are synthetic segments of DNA or RNA (oligonucleotides) designed to mirror specific mRNA sequences and block protein production.

Hybridon has sufficient cash to complete the year, Grinstead said. "We have potential resources from partnership discussions," Grinstead said. *