The Long and Winding Road to FDA Approval of a Novel Prostate Cancer Test: Our Story
Bibliographic record
Abstract
Only a very small fraction of the many tumor biomarkers discovered are successfully translated from the academic research laboratory into clinical practice. The pathway has not always been linear and has involved people with expertise in a wide range of specialties, including basic research, assay development, clinical affairs, regulatory affairs, marketing, business, oncology, and executive-management leadership. The experience with a tumor biomarker, the PCA3 [prostate cancer gene 3 (non-protein coding)] gene, illustrates the long and winding road that must be navigated. Here we reflect, from a historical point of view, on how the interface between academic science, industry, urologists, and clinical laboratories has been essential to advance biomarkers from solid basic science to a validated clinical laboratory test. PCA3 was first described as DD3 in 1999 by Bussemakers and colleagues (1). Researchers in the Isaacs laboratory at Johns Hopkins University used differential-display analysis to compare patterns of mRNA production in benign and malignant prostate tissue, with the goal of identifying unknown genes involved in prostate tumorigenesis. PCA3 expression was further characterized in the Schalken laboratory at Radboud University, Nijmegen (2), and this research confirmed 2 important properties of PCA3 as a prostate cancer (PCa)6 marker: PCA3 expression is prostate tissue specific, and PCA3 is highly overexpressed in PCa compared with benign tissue. This PCa-specific overexpression led the Nijmegen researchers to assess the feasibility of a PCA3-based urine test for PCa detection. In 2001, a license for PCA3 was obtained by DiagnoCure, a company in Québec City, Canada, to develop the assay and provide a clinical application. Transfer of a cancer biomarker to industry requires both the appropriate technology and a proper cancer strategy. The Nijmegen methodology was converted from PCR to nucleic acid sequence–based amplification (NASBA), a method that works at a constant temperature, and then developed as the uPM3 assay. Several medical centers worked with DiagnoCure to assess the potential clinical utility of the uPM3 assay. The first presentation of uPM3 clinical data was made by Drs. Tinzl and Marberger (University of Vienna) at the 2003 American Urological Association meeting (3), and a larger Canadian study led by Dr. Yves Fradet (4) confirmed the potential utility of uPM3 for predicting prostate biopsy outcome. In late 2003, Bostwick Laboratories in Virginia, USA, used the preliminary clinical data to establish a laboratory-developed test that used uPM3 reagents provided by DiagnoCure. During this period, Schalken's laboratory was continuing clinical research that used cell sediment fractions from patient urine samples that were collected after a digital rectal examination (DRE) procedure. The study, published in 2003 (5), established the protocol of a post-DRE urine sample followed by quantitative amplification of PCA3 and prostate-specific antigen (PSA) RNA. This approach is based on the working hypothesis that PCa cells are normally shed into the urinary tract (the DRE facilitates the release of cells and increases the signal). The measurement of PSA RNA in urine is used to normalize PCA3 RNA concentrations to the amount of prostate cells in the urine sample, and the resulting ratio is termed the “PCA3 Score.” In November 2003, Gen-Probe sublicensed the methodology from DiagnoCure to develop the PCA3 test from the prototype DiagnoCure uPM3 assay. Gen-Probe used a transcription-mediated amplification technology similar to NASBA to take advantage of an already available platform (6–8). An immediate challenge was that processing urine samples required centrifugation to remove sediments before testing, a step that would have hindered the test's implementation into clinical practice. The solution was to add urine samples directly to a stabilization buffer, thereby eliminating the centrifugation step and enabling shipment of samples under ambient conditions. It turned out that this sample-handling modification ensured an RNA yield sufficient for analyzing >95% of samples. The combination of the modified sample-handling protocol and the transcription-mediated amplification technology also decreased the imprecision for quantifying PCA3 and PSA RNAs, allowing more accurate determination of the PCA3 Score. Achieving these improvements underscored (and continues to underscore) the importance of the sample-collection procedure and the assay's robustness to the development and validation of any new biomarker assay, not just PCA3. In January 2005, Dr. Leonard Marks from the University of California Los Angeles presented these findings at a Gordon Conference (6). In late 2005, Gen-Probe PCA3 reagents were made available in the US for laboratory-developed tests, and the PROGENSA® PCA3 assay was CE-marked in mid 2006. Although the progression of the assay was from academic research to DiagnoCure to Gen-Probe, many of the activities were taking place at about the same time. Open communication and effective collaboration between all the groups aided in the relatively rapid clinical application of PCA3. Throughout the process, discussion of the potential uses of PCA3 and feedback from PCa specialists and clinical laboratories were instrumental in designing clinical research studies and guiding assay development, as well as the sample-collection protocol. Why perform clinical research studies? A faster path to US Food and Drug Administration (FDA) approval would seem to be a pivotal clinical trial designed for FDA approval. Especially for oncology, clinical research studies are absolutely essential for understanding the clinical and assay requirements for any new biomarker. The goal of these preclinical studies is to define the target population, investigate sample-collection parameters, and identify the proposed intended use so that the FDA clinical trial is properly designed. Thus, a large amount of effort is required to characterize clinical performance and to determine the optimal intended use for an FDA clinical study. The development of collaborative relationships between clinicians and opinion leaders facilitates the collection of valuable information about unmet clinical needs and how the marker might fill those needs. These studies need to have carefully designed protocols and frequent interactions with the clinical study site. To date, multiple peer-reviewed publications have resulted from such studies (9–16), which have shown overall that PCA3 Scores are correlated with the probability of a positive biopsy, the PCA3 Score is independent of prostate size, and lower PCA3 Scores are associated with low-volume/low-grade cancers. Men with one or more negative prostate biopsy results represent a well-known clinical dilemma. Because of the limitations of transrectal ultrasound–guided biopsy, the fear that cancer might be missed often leads to repeat biopsies. Prostate biopsy is associated with discomfort, anxiety, bleeding, and the potential for serious infection requiring hospitalization. Furthermore, the predictive accuracy of serum PSA is diminished in this setting, so there was a clear unmet need for new tests to help both clinicians and patients make more-informed decisions regarding repeat biopsies. On the basis of the clinical research studies and the clinical need, this intended use was targeted for the FDA clinical trial. The many clinical research studies provided essential information about potential clinical uses for the urine PCA3 test. The extensive data and feedback from clinicians became the foundation for early discussions with the FDA. Because PROGENSA PCA3 was a novel test for PCa, premarket approval and an appropriately designed clinical trial were required. Community clinical centers were selected to better investigate how the assay would perform in practice. The pivotal clinical trial included >500 men with previous negative biopsy results at 14 clinical sites over a 9-month period. The data were analyzed extensively by biostatisticians at several locations, including the FDA. The pivotal study confirmed that the PROGENSA PCA3 assay was reproducible across 3 clinical laboratories and that the PCA3 Score, used in conjunction with other clinical information, could help clinicians and patients make more informed decisions about repeat biopsies. As Table 1 shows, it took about 14 years from the 1999 basic-research publication to the 2012 FDA approval. And the long and winding road continues: Ongoing research on PCA3 and other urinary markers will likely lead to other clinical applications in the future. a RT-PCR, reverse-transcription PCR. a RT-PCR, reverse-transcription PCR. prostate cancer nucleic acid sequence–based amplification digital rectal examination prostate-specific antigen US Food and Drug Administration.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.048 | 0.088 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.006 | 0.012 |
| Scholarly communication | 0.018 | 0.032 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.038 | 0.050 |
| Insufficient payload (model declined to judge) | 0.013 | 0.007 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".