Strategic analytics and forecasting in oncology and rare diseases go beyond numbers, as each number is a patient, and each patient is a life. Oncology is a complicated but important pharma category to forecast with oncology assets becoming an increasingly critical part of company portfolios.
Business strategies, forecasts and analytics are used for research and development, while new product portfolio planning requires different methodology and outputs than other functional areas. In these types of forecasts, we need to adjust for risk and uncertainty in the forecast as the product may not be launched yet.
The key challenge to forecasting is to create a process where the needs of function can be met without compromising the integrity of the forecast approach. It is also important to note how we could create the “one source of truth” across an organisation while having no historical sales data.
When looking at the field of oncology or rare diseases, one gold standard approach is the epidemiological method. Forecasters often start with an epidemiology-based approach, using data and assumptions around prevalence, persistence, compliance, and market share to determine how many patients are taking a drug, and use this to forecast future revenue. This model is used when a product is new to the market, or where patterns of usage are complex (such as rare diseases or cell and gene therapy in oncology).
A forecaster should be equipped to bring and validate different data sources to arrive at meaningful insight. These insights can be useful for building a commercial strategy since they establish a deep and more causal relationship between patients and resulting commercial sales. At the end of the day, forecasters are trying to stay ahead of the curve and steer the organisation and investors in a win-win direction.
However, getting detailed in this space is not always easy. The populations for these treatments are often very specific: many drugs are approved for a certain tumour type, specific line of therapy, and/or only in populations with certain biomarkers. To build an accurate oncology model, forecasters must be increasingly patient-focused and deal with challenges including accurate patient identification, the likelihood of treatment switching and discontinuation, and time on therapy for different patients, as well as pricing changes over time. They must also constantly monitor the rapidly evolving marketplace, where many pipeline drugs mean that standards of care and the competitive landscape can change dramatically in the time it takes to bring a drug to market.
These obstacles force pharma companies to rethink their approach to forecasting in oncology. When projections are misaligned, pharma companies can fail to meet revenue goals, disappointing investors and shareholders. But when forecasters get it right, manufacturers can optimise resources to fit the opportunity and produce plans that deliver innovative medicines to patients who need them.
Oncology is recognised as a challenging area to forecast. According to Precedence research the global oncology market was valued at US$286.04 billion in 2021 and is expected to reach over US$581.25 billion by 2030. The increased importance of forecasting has become even more relevant in recent years, due to the specificity of new-era targeted therapies. These therapies target specific subsets of patients, meaning it is paramount to ensure accurate and robust forecasting to limit over-or under-estimation of product usage.
Here is a list of essential guidance to help business strategists and forecasters navigate the shifting times of oncology treatments and build an effective forecasting approach.
1. Patient segmentation
As oncology treatments become increasingly targeted to certain patient populations, best-in-class oncology forecasts require forecasters to split the population into smaller and more specific segments. For example, for second-line therapy in metastatic lung cancer with improved efficacy for a certain biomarker, a forecaster may need to model eight or more segments in order to understand their potential patient population. While smaller segments are often more accurate, each new ‘category’ of segmentation multiplies the forecasting effort.
2. Monthly forecasting model
Before building an oncology forecast model, it is important to understand the level of data granularity that users demand on an immediate and mid-to-long-term basis. Annual models, albeit easier to build and maintain, do not answer key business questions like monthly sales. It can also be difficult to adapt to an event like a data readout, where changes in forecasting output are needed at a monthly level by business users.
These challenges make annual models inflexible with low precision. On the other hand, a monthly model can offer an ideal time granularity for forecasting because it incorporates oncology-specific dynamics based on available monthly data.
3. Identify the right target patient pool
Forecasting in oncology is different from other therapeutic areas because of the significant need to follow patients through different stages, lines, and treatments as they progress through the disease. As important as this is to do, inaccurate identification of the target patient pool has been a common pitfall in oncology forecasting.
Forecasters should split the population into smaller and more specific segments, and accurately model them based on incidence, recurrence, diagnosis, treatment, and other important factors to maximise the accuracy of forecast outputs.
4. Integrate dynamic patient flows
Forecasters must be able to model patients through the different stages of the disease as cancer therapy models have become more complex. They need to assess the advancement of each patient segment, understand how patients move between the lines of therapy, analyze dosing regimens, rates of progression, remission and discontinuation, patient dependency on old and new drugs or therapies, and more. A holistic understanding of the disease space, as opposed to a myopic one, is critical for forecasters to model such complex and dynamic patient flows.
5. Understand dosing patterns
Understanding the dosing patterns of your target patient population is crucial before building a forecasting model. Each of these inputs needs to be modelled differently because patient segmentation and the associated granularity heavily depend on drug dosing specifications. Duration of Treatment (DoT), persistency curve, and cohort models can be considered for oral targeted therapies that carry a fixed dosing approach.
6. Estimate bolus demand
In addition to estimating the first-time users of your new product, forecasters must also look at two additional streams where patients could potentially flow in from:
- Patients transferred from one product to another midway before completing the line of therapy.
- Patients re-initiating the treatment after temporarily discontinuing it due to tolerability issues. This is called the Bolus demand, and it forms a critical part of any forecasting model.
For those key questions, it is important for the strategic forecaster to have a concurrence with the wider cross-functional team so that there is one number across the organisation.
References available on request.
Sanobar Syed is a subject matter expert in the field of pharmaceutical business strategy and forecasting in North America. She has a successful career spanning over 14 years with top global pharmaceutical firms, including Beigene, AbbVie, Novartis, and McKesson. With a master’s degree in Organic chemistry coupled with MBA in marketing, she has established and successfully led strategic forecasting and business analytics and excellence across geographies for multi-million-dollar brands across distinguished organisations.