Need To Clean Up Your Clinical Trials Budgeting And Forecasting Process? Try MOP-FACE

By Chris Chan, IGM Biosciences

Years ago when I was an adorable young lad, a senior company executive told me: “The budget for the upcoming study? Easy peasy. Fifty thousand per patient. All in. Just like the previous study.” Given my affection for all things painless and easy, I decided on the spot to make this a career. Unfortunately, my subsequent journeys into Dante’s Inferno taught me that clinical trials budgeting is appreciably more complex than originally rumored, and pitfalls are as plentiful as Canadian snowflakes and CRO change orders.

Nevertheless, given the outsized importance of clinical expenses, the pursuit of reliable and accurate study budgets is an exceedingly worthy goal. Having devoted significant blood, sweat, and tears (along with some Pink Floyd and Jefferson Airplane) to this effort over the years at multiple biopharmas, I have identified several very useful areas of focus that will substantially improve your clinical budgeting and forecasting efforts.

Like the CTM who drinks CDER at the SPA without getting anyone SAD or MAD, our industry enjoys a good acronym. As such, here is the acronym to remember that will lead you to budgeting enlightenment: MOP-FACE.

Here are the components:

• Multi-variables
• Outsourcing
• Probability of technical success (PTS)
• Full-time employee (FTE) rate
• Accruals
• Change
• Exuberance

Now let’s explore each of the ingredients.

Multi-variables

If I ask: “How much does a new car cost?”, you might respond: “What kind of car? What brand and model? Electric or gas? With or without a sexy spoiler?” Similarly, clinical trials come in a variety of models and trims. Having a firm understanding of the many cost-driving variables is essential to generating or deciphering a good budget. Some significant variables include: number of patients; number of investigator sites and countries; study duration and number of treatment cycles; enrollment rate; frequency of monitoring visits; complexity and volume of tests; drug manufacturing requirements (biologics vs. small molecule? required comparator drugs?), and many more. Whether you use an external service provider’s database or your own internally developed benchmarks, accounting for these variables is paramount. The greater the likelihood of multiple study changes, the more important it is to underscore your assumptions. Not only will this help you generate better budgets, it also will help you explain those pesky budget variations. The simplistically aforementioned “fifty thousand per patient” can be turned into a useful tool. Since key leaders and personnel are already tracking to this, you can use it as the baseline benchmark comparator. To wit: “The new study is twice the cost because it targets twice the number of sites across more countries, expects a much longer enrollment period, factors in more months of treatment cycles, and requires more CT scans for every patient.”

Outsourcing

Quite often when study costs are contemplated, people are focused on external costs such as CRO fees, investigator grants, central and analytical labs, consultants, and so on. However, a big part of the overall cost equation are internal FTE resources. Most or all CRO services can theoretically be done by a sponsor hiring in-house employees instead. Sponsors simply choose to strategically outsource portions of the study work for speed and financial flexibility reasons.

From a quantitative standpoint, my typical rule of thumb is that on a unit or hourly basis, outsourcing is more expensive than hiring in-house. This makes intuitive sense: all else equal, a CRA would command a similar salary whether he/she works for a sponsor or a CRO, and the CRO would want to charge a profit margin on top. However, certain conditions may confound this thumb-rule. For instance, in-house employees on fixed salaries may work more hours than normal without incurring incremental costs, whereas a CRO would typically charge for every incremental unit. On the other hand, if studies progress slower than expected and there isn’t enough work to keep all in-house personnel busy, the cost advantage reduces or even reverses since idle employees incur unproductive fixed costs. Additionally, if the CRO possesses less expensive overseas infrastructure and personnel, they could theoretically pass on savings to the sponsor that could exceed the sponsor’s internal hiring cost advantages.

The punchline is that both external and internal study costs must be viewed holistically, and the outsource/in-source mix may have material cost implications that need to be flushed out and properly calculated during the budget process.

Probability Of Technical Success (PTS)

PTS stands for “probability of technical success.” In drug development, it refers to the probability that a drug candidate will successfully move from one clinical stage to the next. For example, if a drug in Phase 1 is projected to have a 50% probability of going to Phase 2, that drug is said to have a 50% PTS to move to the next stage. Let’s take the following hypothetical odds for a Phase 1 drug:

• 50% chance of moving from Phase 1 to Phase 2
• 50% chance of moving from Phase 2 to Phase 3
• 50% chance of moving from Phase 3 to NDA/BLA filing
• 80% chance of receiving FDA approval after the NDA/BLA is filed.

The approval PTS for that the drug candidate would be 10% (50% x 50% x 50% x 80%). Industry folks also refer to this as LOA, or “likelihood of approval.”

What does this have to do with budgeting? First, PTS should always be considered when generating your budgets and long-range plans, especially if you have a high number of ongoing studies. For instance, let’s suppose you have 10 Phase 2 studies scheduled to complete during the early part of next year. If your budget assumes that all 10 will move into Phase 3 later in the year and only five do (PTS = 50%), the Phase 3 portion of your budget would be overstated by 100%.

Second, if you do incorporate PTS, you need to choose your probability percentages wisely. There are many resources available (both paid and unpaid) that provide historical clinical development PTS by phase, disease areas, and time period. You can use these resources and choose the most appropriate PTS for your drug candidates, or you can generate probabilities internally based on your company’s own experience and expectations. Keeping close track of your selected PTS percentages and the rationales behind them will come in very handy when defending your proposed budget and for explaining budget variances going forward.

Finally, ensure that you “operationalize” PTS-adjusted budgets appropriately when it comes to partnering with the clinical stakeholders. If a $20 million clinical trial is PTS-adjusted by 50%, the trial budget is a net $10 million. However, the actual spend will be either $0 or $20 million. Giving the clinical team a $10 million budget will only cause confusion and consternation. There should be delineation between how the studies are budgeted in aggregate (with PTS adjustments applied) and how individual study budgets are managed operationally by the teams (with no PTS).

Full-Time Equivalent (FTE) Rate

When it comes to forecasting internal resources, you can make the overall effort appreciably more efficient by using higher level full-time equivalent (FTE) rates rather than calculating by specific positions. Rather than costing out dozens or hundreds of incremental FTEs by position and function, follow the 80-20 rule and assume an average FTE rate for each position. A great way to select an FTE rate is to analyze the actual salaries of your company over the past year(s) and add an inflation factor. You can opt for an FTE rate based on all FTE salaries for the entire company for maximum efficiency, but the trade-off would be lower precision (for instance, the egregiously underpaid finance staff would drive the mean rate down to unrealistic depths…). Alternatively, you can generate FTE rates by function and department for increased precision, say by clinical as a whole or by clinical operations, regulatory, biostats, data management, and so on.

One notable thing to keep in mind is that different companies might define “fully burdened FTE rate” differently. For instance, some might use salary plus benefits only. Some might add additional layers such as travel, training, IT-related costs, and lab supplies. Still others might layer in a share of the building lease and utilities, and so forth. The point is that not all fully burdened FTE rates are created equal. Additionally, factor in any unusual circumstances that may have affected any expense category over the relevant years. For example, if your rate incorporates travel expenses and you calculate the rate using the initial COVID-19 pandemic years, your FTE rate will be understated. My recommendation is for you to calculate several FTE rate versions consisting of different expense levels, adjust for circumstances as needed, and then choose the rate or rates that your organization is most comfortable with.

Alternatively, there is another simple way to derive FTE rates: peruse your CRO contracts and use the standard FTE rates specified within these contracts. You can take a composite average of different rates within the same contract or rate card, or even average rates from multiple CRO’s respective contracts. However, in doing this you should be cognizant of possible differentials. For instance, if your company is located within a particularly expensive region (say the biotech belts of Northern California or New England), using CRO rates derived from a more geographically diverse labor force would likely cause you to understate your own internal costs.

Accruals

If you are familiar with Frodo and Sam’s long, perilous journey returning a ring to Mordor, you’ll have a sense of what your finance department encounters on a monthly basis. Financial accruals refer to the process of estimating all expenses incurred by your company over a given period before all of the invoices have arrived. This endeavor is especially challenging for clinical trial expenses, because there are so many associated entities and moving parts to account for. Because it isn’t feasible to contact every relevant CRO, investigator site, central and analytical lab, IRB, translational service, and consultant for a timely estimate every single month, most clinical finance departments utilize specific models and methodologies to generate reasonable estimates. For example, some bigger companies might use a “straight line methodology” (taking the total contract estimate and dividing equally by expected number of months), while others may allocate expenses based on predetermined cost-driving variables such as number of patients enrolled and sites initiated. Saliently, there are no generally accepted standards when it comes to clinical trials accruals. As such, different companies do clinical accruals in a variety of different ways. The reason this matters for budgets? You won’t generate an accurate budget if your projections are inconsistent with how your company accrues expenses.

Suppose you budgeted your study according to expected milestone payments. For a $10 million work order, you expect to make a 50% milestone payment in January of this budget year and 50% in January of the following year. As such, you budget $5 million this year and $5 million next year to correspond with the payments. However, if your financial accountants accrue on a straight-line basis and are informed that all associated work will be performed between January and October of this year, they will expense $1 million per month over that period and accrue the entire $10 million by October. At the end of this year, you would “overspend” your budget by $5 million, or 100%.

Therefore, it is crucial that all relevant personnel — clinical, finance, and senior leadership who review financial reports — get properly trained or at least briefed on the accrual methodologies used by the company. Keep in mind that companies typically use more than one methodology. Some may use very simple models for small Phase 1 trials but exceedingly complex models for larger Phase 3 global studies. Any conceptual disconnect between the folks generating budget projections and the numbers crunchers accruing expenses can result in major budget variances.

Change

At the end of “Top Secret,” Hillary Flammond ruminates: “Things change, people change, hairstyles change, interest rates fluctuate.” In the clinical world, timelines change, protocols morph, enrollment rates and CRA salaries fluctuate — and CROs feed their change order addiction. The constant tidal wave of changes makes clinical budgeting quite the rocky task. While there isn’t much we can do to avoid changes, we can take steps to minimize their stress and impact.

One essential implementation is to generate and maintain detailed lists of budget assumptions. This list should not only include big ticket components like number of patients, site characteristics, and treatment durations, but to the extent possible also more granular variables such monitoring visit frequency, data review pages, number of screen fails, and so on. A great source is your CRO contract’s budget grid, which typically spells out assumptions in detail. What I usually do is maintain at least two rolling assumptions summaries: a “formal” list of higher-level, big-ticket assumptions that I share with senior management and also a list of detailed assumptions that are as granular as possible. As reforecasts occur, I save assumptions versions corresponding to each forecast version. Is it a pain in the rear? Absolutely. But you’ll be very happy you made the effort when the changes and questions spring forth like locusts.

There are three major benefits that make this extra effort worthwhile. One, this becomes an invaluable tool to help explain all subsequent “budget vs. actual” variances demanded by the C-suite and project leaders. Second, this also provides detailed bridges and explanations between every reforecast and change order. Lastly, it enables an indispensable data source to help you generate future study budgets. For instance, by analyzing the progression details from initial to final budget to final actual spend, you may consider incorporating reasonable buffers into later study budgets that could cover reasonably expected overages.

Exuberance

When it comes to study assumptions, clinical personnel are typically as optimistic as Charlie Brown approaching a football. All sites across multiple continents will be promptly initiated and will immediately enroll 2.75 patients per month, and LPI (last patient in) will hit on or before the pretty Gantt chart’s target date. If the trial only succeeds in enrolling 20% of the year’s enrollment target by October, it’s still highly likely the remaining 80% will be achieved the final two months… so please don’t reduce our budget dollars! It is understandable why our clinical colleagues are vessels of optimism: they need to be. Their targets are so important to the company’s goals that they must be “all in” and laser focused. Ergo, when a number cruncher suggests reducing their study budgets based on observed trends, the suggestion equates to: “Just admit that you expect to fail on your key objectives.”  

So how can we thread the needle and extract arguably more realistic budget and reforecast numbers without destroying the hopes and dreams of our hard-working clinical colleagues? There are two effective ways.

One good way is for finance to apply a “management judgement” factor on the back end. How this works is that finance will factor in budget adjustments based on their own calculations and experience. Suppose clinical operations requests a $10 million study budget this year and expects to enroll 100 patients by year end. However, finance believes that enrolling 50 patients and a corresponding budget of $5 million is a more likely outcome. Finance will then give ClinOps the full $10 million budget and offset this with a $5 million reduction in a separate account. Importantly, ClinOps still manages and tracks the $10 million budget and are not responsible for the adjustment. Any judgement overspend/underspend will be explained by finance.

The second method is for senior management to explicitly declare that while reforecasts are intended to target more realistic projections, the original budget is still the “allowable spend” for all project and functional teams. This provides clinical with some level of comfort in reducing their budget reforecast. Clinical isn’t declaring defeat on their goals, and they won’t feel penalized from a budget standpoint should they succeed in meeting those goals. I’ve personally found this to be an especially effective way to solicit more realistic reforecasts.

MOP-FACE Is A Must-Add To Our Clinical Trial Lexicon

Clinical trials budgeting can be a prodigiously daunting task. The vast spectrum of variables and constantly changing character can befuddle even the most experienced number-crunching practitioner. Furthermore, like the birds of the heaven and beasts in the field, there are innumerous budgeting methodologies being deployed within the industry to varying degrees of effectiveness. But as Andy Dufresne once said, “Remember, hope is a good thing, maybe the best of things, and no good thing ever dies.” No matter how big a budget challenge you have in front of you, I am hopeful the concepts discussed herein will optimize your efforts and help make your endeavor more efficient and productive. Therefore, for your next budget cycle, envision my MOP-FACE to ensure a more delightful budget journey.

About The Author:

Growing up, Chris Chan wanted to be Bruce Lee or a Jedi Knight. He wound up doing the next closest thing and became a biotech finance professional. Over a span of three decades (or approximately four times Al Capone’s total prison tenure), Chris has worked in biopharmaceutical companies of various shapes and sizes, primarily in the areas of corporate FP&A and Clinical/R&D Finance. He has given numerous conference presentations and written multiple articles on drug development budgeting, financial accruals, and outsourcing. He currently crunches numbers as the vice president of FP&A at IGM Biosciences. When he retires, he hopes to become Bruce Lee or a Jedi Knight.