Biopharma Long-Term Product Forecasting

Robert F. Siegmund, PhD MBA

Managing Director Life Code GmbH

Definition

Biopharmaceutical forecasting is the process of generating product forecasts in the biopharmaceutical industry to support decision making. This is important for all parts of the value chain, from making decisions in research and development to making decisions in the supply chain and in the commercialization of pharmaceutical products. It is also important for valuation of biopharmaceutical companies, whether they are publicly traded or privately held.

Type of Forecasts

First question is what you are forecasting. I distinguish between production or supply chain forecasts and financial forecasts, there can also be forecast of treated patients with a certain drug. If we start with the production forecast, you will typically forecast grams or kilograms of active pharmaceuticals ingredient (API) or number of vials, syringes, tablets or packs. In a financial forecast you will typically forecast sales in local currency and in a consolidation currency like $ or € for a certain time period. In a financial forecast you can also forecast all the different cost elements: production cost, marketing cost, development costs etc. Derived from that you will calculate a profit and loss account (P&L) and from that you will calculate Net Present Value (NPV) and Return on Investment (ROI). If you forecast treated patients, the forecast will simply contain your number of patients per time frame.

Granularity

Next question is about the granularity of the forecast. There you have three dimensions: market and product, time and time frame, and finally geography. In market and product, it depends if you forecast a whole market or product class, or do you call out the individual products and molecules. Do you forecast only one molecule or do your forecast all the molecules in the product class or all the products in all the product classes in the market. So, for instance in breast cancer do you forecast only HER2+ inhibitors, do you forecast the individual HER2+ inhibitors or do you forecast all the hormone therapies, chemotherapies, immune checkpoint inhibitor therapies, CDKi and so on. So here the main distinction is between the individual product forecast for instance in this example Herceptin (trastuzumab) and the whole market forecast and all variants in between.

Long-Term versus Short-Term

Next time and time frame. You can do a daily, weekly, monthly or yearly forecast and what is the time frame: short term or long term. Here we distinguish between short term, tactical forecasts and long term, strategic forecasts. Supply chain or production forecasts can be both short term and long term, as well as financial forecasts can be both short term or long term. Short term forecasts are typically up to two years or 24 months time frame and are typically of daily, weekly or monthly granularity. Long term or strategic forecasts are typically of five to 15 years time frame and of yearly granularity

Geography

Regarding geography are different levels of granularity. What we call a level 1 forecast is usually just USA and ex-USA or so-called Rest of World (ROW). Level 2 is broken down by major markets so USA, UK, EU4, Rest of EU, Japan and ROW. And level 3 would be all 30 or 40 countries you need to forecast, forecast individually. Of course, definitions may vary but this is the one I used most.

In-Market versus new Product

And finally, an important distinction: in-market versus new product. In-market product means that you are forecasting a brand or product that has been approved, got reimbursement and access and has been launched already. This means you have a lot of empirical data about the performance of the product, and most of all, you have got solid trend data down to a daily level. This makes forecasting so much easier. New product forecast means the product is still in clinical development or at least has not been launched yet, so from a commercial perspective, its an unknown animal, you have no market performance data, no trends, no hard empirical evidence. You have (hopefully) got clinical efficacy and safety data and a lot of market research. From this data you then need to estimate the future sales by year, that is much more difficult. As the physicist Niels Bohr said, “prediction is difficult, especially about the future”. In-market forecasts can be short term (< 24 months) or long term (5-15 years) whereas new product forecasts only make sense as strategic, long-term forecasts

Structure of the Long-Term Forecasting Model

For a new product long-term forecast, where no in-market data is available on the product, one needs to build up the forecast model from fundamentals. The starting point is patients, and the output is product sales, illustrated in Chart 1.

This approach is often referred to as patient-based forecasting. It involves the multiplication of all the factors in the chart. The process begins by identifying the number of patients, followed by an assessment of how many of these are eligible for and have access to treatment. Subsequently, an evaluation is conducted on how many of these accessible and eligible patients receive any form of treatment. Attention is then directed to the specific product, determining its share among those treated patients. Additional variables, such as treatment duration, dosage, and compliance, are also considered, along with the price per dose. These elements are collectively multiplied to estimate sales. This calculation is not a singular event; as the formula suggests, it involves a matrix of all factors (i) across all time points (t). Geographical considerations are also important; this matrix calculation is repeated for each country where a forecast is desired, independently. It is advisable not to amalgamate forecasts across countries but rather to assess each nation individually and then aggregate these results. The rationale is that parameters like patient share are more accurately measured at a national rather than an international level. For instance, a collective patient share for the EU-5 (Germany, France, Italy, Spain, United Kingdom) yields a less meaningful figure compared to a more precise measure obtainable for a specific country, such as Spain, once the product is introduced to the market.

Forecasting Model Inputs

Now let’s move to the next step in the forecasting process, sourcing the model inputs. Refer to Chart 2 for the inputs to the long-term forecast.

Chart 2: Inputs to the Long-Term Forecasting Model

Let’s start at the top of the forecast. Epidemiology will feed into the top box of the forecast. I strongly recommend using prevalence as the base of the forecast, rather than incidence or mortality. You need to understand the prevalence of the disease and predict the future annual growth rate of the disease from published epidemiological data. If there is no good epidemiological data available, like for instance in ultra-orphan disease markets, then you need to do your own primary research to find out. Market Research is key for the following parameters: accessible patients, therapy penetration and product market share. For all these parameters you will need to do quantitative primary market research. A well-established method for estimating peak or equilibrium patient share of a product is con-joint analysis. The core idea is to break down a product into its individual attributes (such as price, color, size, brand, etc.) and then ask respondents to evaluate a series of options where these attributes are combined in different ways. This method helps in uncovering what combination of a limited number of attributes is most influential on respondent choice or decision making. Businesses can use this information to design new products, modify existing ones, or develop marketing strategies that focus on the most valued attributes. Conjoint analysis is especially useful because it mimics real-world buying situations where consumers are often choosing between products that have multiple varying features, and it helps in understanding how these different features influence their decision-making process.

Market Access and pricing will be key to input the %accessible patients for a product or market and of course, a sustainable price for the product. Often dedicated pricing market research is carried out to determine the optimal price point.

Key Events Influencing a Forecast

Regulatory approval is of course the single most important event in a pharmaceutical product forecast, together with the reimbursement approval. Only when both have happened, the full commercial launch can happen. The commercial launch date is the t=0 zero point in the long-term forecast. I typically use an exponential function to then ramp up the product patient share to the peak share depending on market conditions, usually between two to three years. The exponential function I use is a modified logistic growth curve, similar to what is used to calculate bacterial growth for instance (see chart). If the market is more competitive and depends on the share of voice of the product, it will take longer to reach peak share. I was now talking about a level 2 forecast where you model on a country/national level. If you need to model more countries in one, or if the country, like Italy breaks out into different regions with different approval timelines for reimbursement, you will have to go for some kind of composite curve to model the uptake, or you will have to make your forecast more granular, and forecast each country and region separately, basically turn in into a level 3 forecast.

Patent expiry. I see that basically as t=0 for the generics or biosimilars forecast and use the same curve to ramp up the generics or biosimilar (depending on the case) to peak share, usually faster as share of voice less important than price in that situation. To properly estimate the biosimilar impact of course pricing research and biosimilar impact data will be needed, not to make it simply a guess. This biosimilar will then cannibalize the share of the originator product, the share of which will start to decline from patent expiry onwards, in the shape of an inverse logistic curve.

Impact of Clinical Study Results

Data from the clinical trials will be key for the forecast, especially from the pivotal phase III trial. This data, together with your market research data, will determine what you can assume in peak share or equilibrium share for your product. If you are in earlier stages of clinical development, phase I or II, it makes senses to do scenario planning, and plan through different potential outcomes of the phase III and analyze how they would impact peak share of your product, and hence also the financial and supply chain forecast.

Often you have the situation that a study leads to an indication extension or to an improved perception and hence competitive position of the brand in its market. In that case, the forecast needs to be adapted by adapting either the peak share or uptake to peak share in the forecast. In the case of indication extension, it will require a whole new forecast.

Impact of Marketing and Sales Efforts

Welcome to the complicated world of econometrics. I know of no proven, systematic way to calculate the direct impact of marketing and sale investments on the revenue of a pharmaceutical product. For instance, studies in the 1990s have analyzed when a sales territory is vacant, there is still 70%-90% sales turnover in that vacant sales territory. This number is called the “carry-over” rate, and 100% minus the carry-over rate is the sales representative impact. This is just an estimation, of course. With many pharma companies doing away with sales representatives and digital and social media marketing taking over, the direct impact is even harder to calculate. One useful number to reference is the share of voice (SOV). The SOV is defined as your product’s share of the total marketing and sales investment of all competing products added up. If all companies in a competitive market invest about the same, let’s say there are 4 players and each have an SOV of 25% once can assume that there is no impact of the sales investment, and the patient share approach peak share. Hence this share is also called equilibrium share. This is a big topic that would warrant a separate article.

Dynamic Adaptation of the Forecast to Market Changes

How does a forecast adapt to market changes? Most forecasters in pharma still use local spreadsheets as their main tool which are always static, picture in time, kind of models. The world, however, is constantly changing, and the rate of change is accelerating due to technology. As the pre-Socratic philosopher Heraclitus famously said, “you can never step into the same river twice”, because we aren’t the same and the river isn’t the same anymore. With these spreadsheets models what we can do is to have as many assumption cells as possible and clearly documented, and we can then update our assumptions as needed, and regularly. Even a long-term forecast of 10 years forwards should be updated regularly, once new data comes in. I imagine with the powerful Artificial Intelligence and Machine Learning tools emerging nowadays, this will be possible in an automated fashion very soon, so that we will have truly dynamic forecast models.

Recommended Literature

If you want to go into more depth on the Principles of BioPharma Long-Term Product Forecasting, I recommend the following books.

Tennent, J., & Friend, G. (2011). Guide to Business Modelling (3rd ed.). Economist Books.

Cook, A. G. (2015). Forecasting for the Pharmaceutical Industry: Models for New Product and In-Market Forecasting and How to Use Them (2nd ed.). Gower Publishing.