In biopharmaceutical manufacturing today, quality management is critical for steering clear of production problems, capacity bottlenecks, and operation failures. The good news is that manufacturers appear to be doing a better job over the past 10 years.
In our 9th Annual Report and Survey of Biopharmaceutical Manufacturing Capacity and Production , we evaluated, along with more than 80 other biomanufacturing trends, the frequency of batch failures among global biomanufacturers. We weighted information shared by 302 respondents to estimate the batch failure rate for the industry. Based on the responses, batch failures occur on average every 60.3 weeks per facility. This is a significant improvement over last year’s average of 54.5 weeks, and shows a continuing trend over the past five years (Figure 1). Indeed, in 2008, we found batch failures to occur every 40.6 weeks. This means that in five years, the batch failure rate has improved by 49%.
Delving further into the responses, we find some interesting patterns in play. For example, the proportion of respondents who said that the last batch failure at their facility occurred either two years or more ago stands at 36.5% this year, up significantly from 29.9% last year, 25.8% in 2010, and 26.7% in 2009. On a similarly encouraging note, the proportion experiencing a failure in the past one to three months dropped to 14.1% this year, after being steadily around the 20% mark for the past few years (18.5% last year, 21.1% in 2010, and 21.6% in 2011.)
Tempering the good news, though, is our finding that the proportion of respondents experiencing batch failures very recently (within the last week or last month) is markedly up. This year, more than 1 in 10 (10.6%) reported a failure either within the last month (8.2%) or the last week (2.4%). This is a step above the 7-8% who have indicated this in past studies.
Taken together, though, the news on the whole is encouraging. The continuing reduction in frequency of batch failures is a good sign, and represents a maturation in performance, likely even within smaller organizations. Some of this improvement is directly related to training of operations staff, which, according to the study, received significant budget increases this year.
Although the specific causes contributing to this improvement are not fully defined, companies are clearly managing their manufacturing more effectively, most likely by: improving their process design; resolving supply chain issues; using increased process monitoring and process analytical technology (PAT); gaining experience in preventing contamination; and otherwise learning from prior contamination episodes. Also, it is possible that “natural selection” is at work, with those companies experiencing more process failures also tending to have other quality and management problems contributing to failures.
PAT Adoption on the Rise
One potential reason for the decline in batch failure frequency is the industry’s increased adoption of Process Analytical Technologies (PAT). PAT is defined by FDA as “a system for designing, analyzing, and controlling manufacturing through timely measurements (i.e., during processing) of critical quality and performance attributes of raw and in-process materials and processes with the goal of ensuring final product quality.” In many respects PAT is nothing new and involves no new specific requirements beyond those needed to support cGMP approval. PAT, Quality by Design (QbD) and other process measurement-based quality programs are efforts to better quantify, model and otherwise understand manufacturing processes.
Our study shows that continued improvements in sensors, probes and analytical equipment are facilitating process quantification and PAT. Thus, as bioprocessing becomes increasingly monitored by improved and new chemical, physical and microbiological detection methods and assays, including single-use sensors/probes, the resulting data will increasingly support and be used for mathematical modeling and risk analysis. Besides this technological progress promoting increased use of PAT or comparable quality programs, industry adoption will also likely increase as PAT is recognized as an effective method to increase productivity by reducing waste, improving yields, increasing automation and facilitating other cost-saving measures.
Our survey data supports this view. When we asked respondents about the quality initiatives they have currently implemented, just 21.3% cited PAT, the lowest of the 12 initiatives we identified, and far behind others such as QbD and risk analysis. This may not be surprising, given that adoption of PAT is voluntary. However, when we factor in respondents’ plans for the next 12 months, the story changes. Indeed, 29.3% of respondents plan to use PAT in the next year, the highest proportion of any of the initiatives, and up from 16.1% who responded that way last year. This puts PAT adoption on par with process modeling (52% using or planning) and knowledge management (50.6%), and ahead of other initiatives such as multivariate data analysis, factorial testing of critical process parameters, and stage gate and in-line product reviews.
Increased use of PAT may also be owing to the lessening burden presented by various hurdles to implementation. When we asked respondents about the most significant hurdles in implementing PAT, we found that, in general, most factors are on a multi-year decline. For example, the most common factor identified as significant or very significant, “time required to implement,” was cited this year by just under three-quarters of respondents, down from 79.5% in 2009. And while “insufficient people in-house to manage implementation” remains an issue, it also continues a gradual decline.