Growing Agchem R&d

Research in agrochemicals used to be simple: "spray and pray," as one R&D director puts it. But that time is long gone in the $30 billion crop protection industry. Pressures from regulators, shareholders, environmental activists, and consumers are driving companies in the business to product development that increasingly mirrors that of the pharmaceutical industry.

So it is not surprising to see the tools and techniques developed by drugmakers being used by agrochemical researchers. What's more, those researchers say, in many ways they are finding more success with the same tools than their counterparts in human health have.

That's good news for the agrochemical industry. Companies in the sector are counting on R&D to keep their pipelines of new products well-filled because, just as for drug companies, a full pipeline is key to the health and growth of the agchem industry.

For example, Bayer CropScience told investors earlier this month that it has doubled its forecast for sales of crop protection products containing new active ingredients. Previously, the target for sales of such products had been set at 1 billion euros--roughly $1.2 billion--per year, to be achieved by the end of 2006. Now, says Friedrich Berschauer, chairman of this Bayer company, the company's target is 2 billion euros by the end of the decade.

Similarly, BASF's Hans W. Reiners, president of its agricultural products division, sees new products contributing sales of $2.2 billion by the end of the decade.

FOR AGROCHEMICAL companies, research into new molecules means a substantial commitment. Bayer figures that it costs nearly $250 million from inception to launch of a new product, often 10 years later--up from roughly $60 million in 1990.

Spending on initial chemistry is roughly $80 million. Biology work, begun soon after promising chemistry results, costs nearly the same amount. Follow-up toxicology and environmental analysis adds another nearly $85 million to the tab. Only after eight or nine years can a company submit registration documents to regulatory authorities. And only after regulators clear the product can it hit the market and begin to recoup those costs.

For researchers at the six major agrochemical companies-- BASF, Bayer, Dow Chemical, DuPont, Monsanto, and Syngenta--there are several fundamental differences between their industry and the pharmaceutical industry.

The big difference--and advantage--that David Lawrence, head of research and technology at Syngenta, points out is: "We can test directly for the effect we want. We can test immediately on plants, fungi, and insects. You can't do that in pharma."

Moreover, notes Bernward Garthoff, chief scientific officer for Bayer CropScience, agrochemical researchers get to double their fun: They can use the two hemispheres to pack two seasons' worth of research into one year.

On the other hand, Peter Eckes, BASF's senior vice president for global research in the agricultural products division, points out that drug researchers concentrate on the action of a substance in the human or animal patient with absorption, distribution, metabolism, excretion, and toxicology (ADME/Tox) studies. Agrochemical researchers, however, must go one step further into the environmental impact of the substance.

Within the framework of those fundamental differences, however, there actually are many similarities, Garthoff says. "When I moved from pharma to agrochemicals, the surprise was that it was not that different," he recalls. "I spent 18 years in pharma and almost 12 years in agrochemicals. They are quite similar; the basics are the same."

Dan Kittle, vice president of R&D at Dow AgroSciences, agrees. "Research strategies are virtually identical," he says. "We use the same tools, but on an ag budget."

However, as Kittle points out, when and if a pharmaceutical active ingredient is successful, the company developing it can enjoy tremendous success. "The investment/reward ratios are different," he explains. "But that's understandable. I'd pay a lot more to have my daughter treated than my field of soybeans."

THE TRUE TEST of the closeness between agrochemical and pharmaceutical R&D, Kittle adds, comes down to staff. "I compete with the pharma folks toe-to-toe for chemists, biochemists, molecular designers, and so on."

Pat Confalone, DuPont Crop Protection's vice president of global R&D, believes that his industry's main change in recent years has been the increased use of new discovery platforms. High-throughput screening remains a way of finding leads, but other approaches include combinatorial chemistry, directed library designs, patent and literature scouting, natural product extracts, and structure- or mechanism-based design.

Any survey of the agrochemical industry's greatest need, Confalone says, will arrive at one conclusion: "Like any other R&D-intensive business, we need a great lead compound--that's the beginning of any dedicated discovery program. If you look back 10 years, it was leads. If you look back 20 years, it was leads. So the major discovery platforms employed today are all designed to identify new lead compounds."

High-throughput screening, however, does have some drawbacks, says Lawrence, whose career wends back to ICI, which owned the Zeneca agrochemical business that became part of Syngenta. At ICI, he recalls, "We started high-throughput screening in 1983. At the time, we were doing more in agrochemicals than in pharma. Then genomics came in, and combinatorial chemistry, and automation, which enabled us to do more."

"Six or seven years ago, we would have said we were going into the exploitation of this enhanced high-throughput screening technology," Lawrence says. "But we had less money to spend on it than on pharma, so we couldn't go for the big numbers."

By Lawrence's reckoning, though, that was okay. "In our experience, chemical library technology is quite restricted to the kind of molecules you can make. You do a skeleton and a lot of analogs, but the diversity isn't all that big. There is a huge potential, and libraries only scratch the surface." Syngenta responded by stepping away from really high throughput to about 1 million compounds per year, and he says, "We've stepped back even from, say, half a million. Throughput occupied our time but didn't give us product."

Instead, Syngenta has put more emphasis on designing molecules for the effects its researchers want. "We are using genomics to understand what is happening when we get the hits. We do that through automation and whole-organism tests. That seems to be working for us," Lawrence says.

The trick, he adds, "is to find a molecule with an edge that can replace something that has been around for years." It should, he says, be an improvement in basic factors such as activity, potency, and safety. But it should also be an improvement in convenience, fitting what a customer really wants.

At BASF, Eckes says, miniaturization of tests is helping screen more compounds and make decisions faster. This has boosted the throughput of in vivo organism-based screening by a factor of 10. It has also enabled the company to look at more diverse compounds. One result: BASF claims that some 70–80% of the compounds it decides to pursue will make it to market.

Eckes is also enthusiastic about in vitro technology. "We are looking at genomics to identify new modes of action," he says. "In this respect, crop protection probably is more advanced than pharma. We can do targeted creation of antisense plants--introducing antisense genes in a plant, gene by gene. You can down-regulate one gene and look at the effect on the plant."

A successful new product in the pharmaceutical industry has a short, fast, high peak of profitability, then abruptly drops off as it comes off patent, Garthoff points out. Agrochemicals, on the other hand, follow a shallower, wider curve. "You never have the sharp peak," he explains. "We have a longer penetration period. We have to cover all the different crops, for example."

And that is why, he adds, a producer tries to identify as early as possible a new mechanism specific for one family of insects, perhaps, but extending to several. "The broader the spectrum, the better," he says.

One area of significant difference between agrochemical and pharmaceutical research remains the frequency of collaborations with outside companies and institutes.

In agrochemicals, Lawrence says, there is not much outlicensing by small biotechnology or active-ingredients discovery companies. "There just are not that many people doing it in crop protection. They are more into pharma. There, if they hit it big, they hit it really big." One exception is gene technology, where he sees a lot of small companies studying genetic traits.

"Any biotech company that is interesting to the pharma industry is also interesting to us," Garthoff says. Bayer has made some investments in start-up companies in agbiotech. And it has formed a joint venture with biotech specialist Exelixis in insects and insecticides. "Exelixis normally is more human biotech, but our interest is insects," he says. "Perhaps we will find some substances--new targets, new mechanisms--through this venture that will help us in the early phases of our research."

Dow has also established some external agrochemical R&D alliances, according to Kittle. In fact, he says, external R&D is more than a third of his agrochemical R&D budget.

This May, for example, Dow agreed to a $10 million strategic alliance in plant biotechnology with the National Research Council Canada's Plant Biotechnology Institute in Saskatoon, Saskatchewan. It builds on a previous $10 million pact in the area of canola, including plants that produce higher levels of oil and canola meal with reduced levels of antinutritional factors.

Some of the most recent R&D efforts have come down to companies doing more with what they have.

For example, Eckes says his colleagues "have made some interesting observations that some of our products--especially the strobilurin fungicides--not only are good at combating diseases but also have a favorable effect on crops. Crops stay green longer, and they ripen a little later, so a plant has time to produce more proteins and carbohydrates. These products somehow make the plants more vigorous and increase yield," Eckes says.

PRODUCT FORMULATION also involves finding new delivery techniques, Lawrence points out. A case in point is the increasing importance of seed treatment. "Rather than spraying a chemical over a crop and the ground, it is much nicer to just have the coating on a seed," he says. BASF, for example, recently inaugurated a new seed treatment technology center at its Limburgerhof, Germany, site.

Most of the primary research in agrochemicals is done by the so-called big six, but smaller or more specialized companies are involved as well, especially in generating research data to support new products. And those companies are becoming increasingly squeezed by rising costs.

One of the latest battlegrounds is the European Union, where authorities are requiring that pesticide active ingredients currently on the market undergo reevaluation. But part of their proposed regimen includes the obligatory sharing of registration data with other producers, a point that worries small- and medium-sized producers in particular.

In fact, a coalition of these producers was formed in June to explain to European Commission officials that the new regulation, as proposed, would seriously erode data protection, effectively nudging smaller firms out of R&D in the chemical crop protection sector.

Euros Jones, senior regulatory affairs manager at the European Crop Protection Association (ECPA) in Brussels, says the EC proposal would effectively punish companies investing in R&D that supports product registrations. Currently, he points out, if one company makes the investment to register a generic product or formulation, other companies must compile equivalent data or contract or license the data.

The proposal will not apply to new products, only to off-patent active ingredients. But generics producers are taking an increasing share of the European agrochemical market. In 1999, the big multinationals held 86.7% of a $10.4 billion market, compared with 8.3% for major generics producers and 5.0% for the coalition companies. Last year, however, the large firms accounted for only 81.3% of a $10.9 billion market, while the major generics producers had 11.7% and the coalition companies held 7.0%.

Registration costs are of particular concern for makers of pesticides used on vegetable or fruit crops, an important sector for European agriculture. According to ECPA statistics, some 75 active substances are used in the EU for fruits and vegetables with annual sales of less than $25 million each--three-quarters of the total substances currently authorized for use in this agricultural sector. According to ECPA, the coalition members are the main data submitters for products incorporating any of more than 30 active substances.

BUT EVEN LARGE companies are feeling the registration cost pinch. Reiners says companies will look at concentrating on fewer, more profitable products and possibly eschew products for specialty crops or for just one country.

For Kittle, many of the changes that have occurred in the world of agrochemical research reflect changes in the agricultural industry itself. "The agricultural industry's mission has changed over the last 10 years," he says. It used to be focused on increasing the world food supply. But now it also focuses on the quality of the food supply, for example, the healthy oils program promoting high-oleic, no-trans-fats oils. Moreover, the industry is coping with dramatic changes in energy supply, demand for renewable feedstocks, development of biomaterials, and so on.

"The agricultural industry used to focus on the grower," he adds. "Now there are also the regulators, the food chain, and its end consumers. The consumers are asking, 'What do we want to buy?' and it's not necessarily the same as what is produced. The consumers are asking what agricultural products are used, where, and in what dosages." They are making decisions that directly feed back to the rest of the food chain, he notes.

Kittle adds: "We're being buried in data. We need some kind of transformation to cut down the information--a new computational tool to translate real life into in silico models, perhaps. We need to know better formulation function, the interdependence of proteins, and targeted delivery systems--not just delivery onto a plant, but stewardship as well."

As Confalone notes, there has been no one discovery platform that the industry has settled on to the exclusion of all others. "We must employ all the tools in the tool kit." He argues that R&D still needs to understand penetration, absorption, and so on--basically, everything that happens when an agrochemical goes on a crop or pest. This information then assists in the modification of the chemotype in order to improve potency, spectrum, and selectivity.

Most of the companies doing R&D "will set the dial differently on the various discovery platforms; they all have pros and cons," Confalone says. "The goal is innovation--not just innovation in the 'eureka' sense, but in the field, delivering what the customer wants."

That's no "eureka" discovery, just good sense in a down-to-earth industry.