Attend any conference dealing with the pharmaceutical industry, from new drug discovery to API and drug product manufacturing, and you will hear presentations outlining successful programs and stunningly brilliant research and development. These inspiring stories, whether ours or someone else’s, punctuate why we became and remain engaged in one of the most challenging and rewarding endeavors, the discovery of safe and effective new drugs. But the very nature of pharmaceutical research and development presents us with more failure than success. Reactions fail to yield the desired products. Potential new compounds fail in vitro profile screens. Promising candidates present unexpected adverse effects in vivo. Synthetic routes collapse upon scale-up. In reality, more things go wrong than as planned.
So when we engage a contract research organization (CRO) or a contract manufacturing organization (CMO) and execute a plan of action, why should we expect anything different?
Plan to Succeed, Prepare for Failure
Design of a project to be outsourced must include extensive scenario development. While attempting to plan for every possible outcome will quickly become a Quixotic quest, careful evaluation of the key aspects and the impact on the project if these key results are not achieved will quickly prioritize which scenarios are most likely. This extra effort is critical to the success of the outsourced project because project flexibility will always be diminished in the outsourcing arena. When you and your contractor are well prepared for disappointing results, the negative impact on the project will be minimized
The relationship between the client and the provider also affects how negative results impact the project. Mutual trust and respect will eliminate non-productive behavior and promote problem solving activities. Transparent effective continuous communication across all levels between the client and provider is critical to the planning as well as to the problem resolution. Nothing solves a technical issue better than direct communication between the scientists actually doing the work.
Realistic Expectations
A typical research and development project plan will incorporate aspects such as desired outcome, primary and secondary strategies, and required resources usually quantified as time and materials. One aspect that cannot be overlooked is risk assessment. Risk assessment can be a complex task and a while a thorough discussion is beyond the scope of this article, when a project plan is designated for outsourcing, realistic expectations must be supported by competent risk assessment. A CRO repeating a known synthesis is much more likely to proceed without mishap than the CRO developing a new synthesis of a proposed scale will likely encounter fewer missteps than using a procedure that only prepared 50g. These are simple examples but they demonstrate how expectations must be in line with the challenges of the project.
Whatever issues exist must not only be examined carefully during internal project planning but must also be shared with the service provider. With both client and contractor mutually aware of the risks involved with a given project, a firm foundational relationship is established which will not only serve to minimize the stress of an undesired result but also maximize the collaboration needed to solve the problem. In a real life example, a CRO was given a project and no instructions or guidance save the desired result. They toiled for months with neither success nor communication with the client. Anxiety and frustration grew exponentially because they were a small company and the client was a large pharmaceutical company and they were hoping to make a good impression. Finally, with complete resignation, they contacted the client and told them they were unable to complete the project. The client’s response? “Oh. That’s what the last two CRO’s said”. In stark contrast, my company needed to outsource the synthesis of an API to a CMO and we had an ample timeline. My group performed a 1Kg demonstration run, and we saved out about 100g of each stable intermediate. I wrote the experimental details for the technology transfer document based upon this run, and then sent the document to the CMO along with the intermediates and a 1g sample of the final product. The analytical method was included in the technology transfer document as well. We then set the date for the face to face meeting for three weeks later. By the time we arrived, the chemists at the CMO were able to run most of the steps in parallel on a multi-gram scale, and we sat in the conference room talking chemist to chemist about the actual reactions. The information exchange was tremendous, and by the end of the day, the project manager for the CMO exclaimed that this was the best technology transfer process he had witnessed in his twenty years of experience. The synthesis scaled up to a 10Kg level with minor modifications and the API was delivered ahead of schedule.
Project Management
Maintaining current detailed awareness of an outsourced project requires planning, effort and commitment. Internal project updates usually occur spontaneously, informally and immediately. New and interesting results are shared constantly, and small adjustments in the project direction are made almost continuously. Outsourced project updates are scheduled, more formal and the subtle changes that can keep an internal project fine tuned are often absent in the outsourced project. New information is created at different rates depending on the project; you need to schedule updates so that assimilation of the new data is manageable, but no more than two weeks should pass between meetings. Have the contractor adapt a written report format that fits your needs and let the format evolve to best fit the project and the metabolite. A CMO working from a procedure demonstrated on a 1Kg people communicating. Teleconferences should be held at the same rate of reports and it is critical that the client reviews the report prior to the teleconference. Insure that both the positive results and the failures are equally communicated, and if possible the teleconferences should allow scientist to scientist interaction so that technical matters can be properly addressed. When issues arise, agree upon action items and clearly assign responsibilities. Just because the client is paying the provider to work on the project does not relieve the client of support responsibilities. This could include literature searching, surveying the lab for expertise or providing copies of laboratory notebook pages or analytical data. Design the report style and teleconferences carefully to fully convey as much information as possible. Make the effort to review the reports in order to maximize the effectiveness of the teleconferences and follow through on all items that need attention.
Dealing With Adversity
Things can go wrong at many levels. The most important parts of how you respond to a negative result are your honest objective assessment of the impact that result has on the project and your ability to accurately convey that assessment to the contractor. Your response should be aligned with your expectations as well. When you communicate your response to the provider, it must be honest, transparent and constructive. Overstating the severity of the negative impact in an attempt to motivate the contactor sets poor precedence on many levels. The contractor is already upset with the negative result and any provider worth working with is already motivated to successfully complete your project. And if every little miscue has you screaming in anger, when a bona fide issue presents itself and you really need the contractor to respond, they may have become desensitized and not respond in a manner you expect. Minimization of the negative impact on the other hand sets inaccurate prioritization for the contractor. This mismatch in client/provider expectation will likely lead to further animosity. Accurately communicating the level of negative impact to the contractor and listening carefully to understand the effect the negative result has on the contractor will keep expectations in balance. For example, consider a six step synthesis for a promising new compound that has been delegated to a CRO, and the first step gives a 52% yield when your technology transfer data projected a 78% yield.
Situation 1: On your first teleconference the contractor expresses grave concern and wants to repeat the procedure because they have essentially lost their overage allowance on the first step. Your primary need is more material for further testing and less material than originally targeted will be sufficient. Furthermore, waiting for more starting material would compromise your testing schedule. In this case the negative impact of the poor yield in the first step is not significant. It is imperative that you communicate the situation to the CRO so they don’t waste time with more starting material and get moving forward with the rest of the synthesis.
Situation 2: On your first teleconference the contractor expresses little concern over the low yield and suggests ordering an additional portion of starting material to make up the balance at no extra cost. They will get on with the synthesis and carry along a second batch. Your testing is critical for advancement of the project and all testing must be done on one uniform batch. Although it might set the schedule behind, the CRO must wait until the second batch of step one is complete before moving on. In this case the negative impact is much greater and the CRO must understand completely the situation and how to proceed forward.
These illustrations are fairly simple and straight forward. Negative results in real projects are often more complex and have more than one potential solution. The key principals to take away from this discussion are honest evaluation of the negative impact, clear and accurate communication to the contractor of the severity the negative result has caused, and the best way forward to constructively deal with the situation.
Another situation that is avoided by establishing with the CRO/CMO a strong relationship of trust, open communication and honest coolheaded evaluation of negative results is what I call the “sweep it under the rug” scenario. If the CRO/CMO is uncomfortable reporting negative results for any reason, they may move ahead on their own to resolve the situation so they can report success, and in doing so diminish the overall value of the body of work. As another real life example, we had engaged a CMO to carry out a multiple step synthesis of a rather complex API with multiple chiral centers. The starting material set some of the chiral centers, but certain steps had to be carried out under careful conditions to maintain selectivity when a new chiral center was created. In one such step, temperature control for an overnight reaction was critical in obtaining a ratio of greater than >95 to <5 of the desired diastereomer so that purification could be carried out by simple recrystallization. The reaction controls failed, however, the reaction warmed up and a ratio of about 75 to 25 was obtained. Instead of telling us immediately about the negative result, they divided the approximate 1500g of crude material into thirty lots of 50 grams each and every chemist in the company was assigned a lot and had to purify it by chromatography. By the time we had the next scheduled teleconference, they had combined all the purified material into one batch, ran a quick recytallization to polish it and were ready to move on to the next step. While they eventually succeeded in making enough of the material in specifications and on time for our clinical trials to commence, we lost the opportunity to examine the upset material and to see if a kinetic resolution under basic conditions could improve the ratio, and we certainly did not have a process that was ready for scale-up to provide material for the next set of clinical trials.
Things Besides the Chemistry Can Go Wrong
In addition to the precarious nature of research and development to present you with negative results, non-scientific issues can also plague outsourced projects. The very nature of the protocols your company follows to establish contractual agreements and arrange for payment to contractors can be vexing. Our medicinal chemistry group had been maintaining a FTE arrangement with a CRO in India. At every renewal date, copious paperwork had to be filled out and due to my company’s standard operating procedures and the original contracts had to be signed in duplicate by both parties for execution of the contract. On one occasion a confluence of events including a time consuming typographical error, the contracts sitting on someone’s desk for two weeks during a vacation and an oversight regarding the address on the mailing envelope resulted in the new quarter beginning without the contract in place. Fortunately, I was able to contact all the parties involved and explain the situation. The CRO generously worked for about a month without a contract based primarily on my assurance that the funding had been approved and that the contract was in the mail. Sometimes openly and honestly communicating the negative result starts with the client and not the contractor.
When working with a CMO in Taiwan, we found weekly teleconferences to be inefficient due to the time difference so we modified the reporting structure to weekly reports via secure e-mail and once a month teleconferences. In the middle of the project, communication went dead. For three weeks, there were no reports and all communication channels were ineffective. Finally, a barely legible copy of a facsimile from the CMO reached me by interoffice mail. Apparently there had been a fire on the CMO site. No one was injured, but the server/phone room was a total loss and they were without e-mail and phones. They sent a FAX to a remote location of my company which someone put in the mail to me. Eventually communications were re-established, and we were pleased to discover that the fire had resulted in no negative impact on the project. My management was not pleased when my project updates were not on time, but the eventual explanation was accepted without permanent damage to my career.
When Things REALLY Go Wrong
There will be some instances when, for one reason or a combination of reasons, that a project will not reach the desired outcome. These can be the most difficult situations to manage. Depending upon the language in the contract, you may not be obligated to pay anything to the contractor; however if an honest effort has been put forth, all negative results were suitably communicated and all appropriate efforts to overcome the problems were taken, in my opinion the correct course of action is to compensate the CRO for some or all of the work, depending upon the circumstances. Ensuring a comprehensive report is written that captures all the failed experiments and undesired results has value, especially if additional effort will be made to achieve the desired outcome. In some cases only compensation for materials can be made. These will need to be decided on a case to case basis. Your negotiations with the contractor should incorporate the same open and honest communication that your project reports used, and the long-term relationship between client and contractor is often more important than the dollar value of the failed project at hand.
There will also be times when you no longer seek to work with a contractor. It may be due to a failed project, a change in company business strategy, or a reduction in funds available for designating projects to be outsourced. In this situation, the details should be candidly disclosed to the contractor. If the cessation of activity is not due to poor performance, the CRO/CMO might ask for a recommendation or to use your name in promotional literature. If your company allows such endorsements, by all means comply with their request. If there has been performance issues, clearly communicating these to the CRO/CMO in a constructive and informative manor gives them a chance to learn and improve. The CRO/CMO world is a small one, and with career changes due to layoffs, mergers and acquisitions and job upgrades, it pays to maintain relationships with contractors even through adversity, for you never know where and when and with whom you may be working.
Murphy’s Law states, “Anything that can go wrong will go wrong”. If you spend enough time in the CRO/CMO space, you will become a living breathing proof of this law, and have plenty of stories to share over a drink or two at the next conference, long after the success stories have been told.
Jeff Marra received his B.S. in Chemistry from Clarkson University and his Ph.D. in Organic Chemistry from Colorado State University. After 12 years of industrial process R&D, and 15 years spent managing the outsourcing of research and development compounds, and more recently the outsourcing of API compounds for Purdue Pharma LP, the author is exploring to find the next interesting opportunity