Report of the Pathology Committee: false-positive and false-negative diagnoses of prostate cancer

Validation and three years of clinical experience in using an artificial intelligence algorithm as a second read system for prostate cancer diagnosis-real-world experience

Background

Prostate cancer ranks as the most frequently diagnosed cancer in men in the USA, with significant mortality rates. Early detection is pivotal for optimal patient outcomes, providing increased treatment options and potentially less invasive interventions. There remain significant challenges in prostate cancer histopathology, including the potential for missed diagnoses due to pathologist variability and subjective interpretations.

Methods

To address these challenges, this study investigates the ability of artificial intelligence (AI) to enhance diagnostic accuracy. The Galen™ Prostate AI algorithm was validated on a cohort of Puerto Rican men to demonstrate its efficacy in cancer detection and Gleason grading. Subsequently, the AI algorithm was integrated into routine clinical practice during a 3-year period at a CLIA certified precision pathology laboratory.

Results

The Galen™ Prostate AI algorithm showed a 96.7% (95% CI 95.6-97.8) specificity and a 96.6% (95% CI 93.3-98.8) sensitivity for prostate cancer detection and 82.1% specificity (95% CI 73.9-88.5) and 81.1% sensitivity (95% CI 73.7-87.2) for distinction of Gleason Grade Group 1 from Grade Group 2+. The subsequent AI integration into routine clinical use examined prostate cancer diagnoses on >122,000 slides and 9200 cases over 3 years and had an overall AI Impact ™ factor of 1.8%.

Conclusions

The potential of AI to be a powerful, reliable, and effective diagnostic tool for pathologists is highlighted, while the AI Impact™ in a real-world setting demonstrates the ability of AI to standardize prostate cancer diagnosis at a high level of performance across pathologists.

Isomer-Resolved Imaging of Prostate Cancer Tissues Reveals Specific Lipid Unsaturation Profiles Associated With Lymphocytes and Abnormal Prostate Epithelia

Prostate cancer is the fourth most common cancer worldwide with definitive diagnosis reliant on biopsy and human-graded histopathology. As with other pathologies, grading based on classical haematoxylin and eosin (H&E) staining of formalin fixed paraffin-embedded material can be prone to variation between pathologists, prompting investigation of biomolecular markers. Comprising around 50% of cellular mass, and with known metabolic variations in cancer, lipids provide a promising target for molecular pathology. Here we apply isomer-resolved lipidomics in combination with imaging mass spectrometry to interrogate tissue sections from radical prostatectomy specimens. Guided by the histopathological assessment of adjacent tissue sections, regions of interest are investigated for molecular signatures associated with lipid metabolism, especially desaturation and elongation pathways. Monitoring one of the most abundant cellular membrane lipids within these tissues, phosphatidylcholine (PC) 34:1, high positive correlation was observed between the n-9 isomer (site of unsaturation 9-carbons from the methyl terminus) and epithelial cells from potential pre-malignant lesions, while the n-7 isomer abundance was observed to correlate with immune cell infiltration and inflammation. The correlation of lipid isomer signatures with human disease states in tissue suggests a future role for isomer-resolved mass spectrometry imaging in assisting pathologists with prostate cancer diagnoses and patient stratification.

Autoregressive parametric modeling combined ANOVA approach for label-free-based cancerous and normal cells discrimination

Label free based methods received huge interest in the field of bio cell characterizations because they do not cause any cell damage nor contribute any change in its compositions. This work takes a close outlook of cancerous cells discrimination from normal cells utilizing parametric modeling approach. Autoregressive (AR) modeling technique is used to fit the measured optical transmittance profiles of both cancer and normal cells. The transmitted light intensity, when passes through the cells, gets affected by their intercellular compositions and membrane properties. In this study, four types of cells: lung-cancerous and normal, liver-cancerous and normal, were suspended in their corresponding medium and their transmission characteristics were collected and processed. The AR coefficients of each type of the cell were analyzed with the statistical technique called Analysis of variance (ANOVA), which provided the significant coefficients. The poles extracted from the significant coefficients resulted in an improved demarcation for normal and cancer cells. These outcomes can be further utilized for cell classification using statistical tools.

Highly Selective Optical Sensor Eu (TTA)3 Phen Embedded in Poly Methylmethacrylate for Assessment of Total Prostate Specific Antigen Tumor Marker in Male Serum Suffering Prostate Diseases

A low-cost, simple, and highly selective method was used for the assessment of total prostate specific antigen (tPSA) in the serum of prostate cancer patients. This method is based on quenching the intensity of luminescence displayed by the optical sensor Eu (TTA)₃ phen/poly methylmethacrylate (PMMA) thin membrane or film upon adding different concentrations of tPSA. The luminescent optical sensor was synthesized and characterized through absorption, emission, scanning electron microscopy (SEM), and x-ray diffraction (XRD), and is tailored to present red luminescence at 614 nm upon excitation at 395 nm in water. The fabricated sensor fluorescence intensity is quenched in the presence of tPSA in aqueous media. The fluorescence resonance energy transfer (FRET) is the main mechanism by which the sensor performs. The sensor was successfully utilized to estimate tPSA in the serum of patients suffering prostate cancer in a time and cost effective way. The statistical results of the method were satisfactory with 0.0469 ng mL^-1 as a detection limit and 0.99 as a correlation coefficient.

Errors in prostate core biopsy diagnosis in an era of specialisation and double reporting

AIM

To examine the effects of specialist reporting on error rates in prostate core biopsy diagnosis.

METHOD

Biopsies were reported by eight specialist uropathologists over 3 years. New cancer diagnoses were double-reported and all biopsies were reviewed for the multidisciplinary team (MDT) meeting. Diagnostic alterations were recorded in supplementary reports and error rates were compared with a decade previously.

RESULTS

2600 biopsies were reported. 64.1% contained adenocarcinoma, a 19.7% increase. The false-positive error rate had reduced from 0.4% to 0.06%. The false-negative error rate had increased from 1.5% to 1.8%, but represented fewer absolute errors due to increased cancer incidence.

CONCLUSIONS

Specialisation and double-reporting have reduced false-positive errors. MDT review of negative cores continues to identify a very low number of false-negative errors. Our data represents a 'gold standard' for prostate biopsy diagnostic error rates. Increased use of MRI-targeted biopsies may alter error rates and their future clinical significance.

Novel artificial intelligence system increases the detection of prostate cancer in whole slide images of core needle biopsies

Prostate cancer (PrCa) is the second most common cancer among men in the United States. The gold standard for detecting PrCa is the examination of prostate needle core biopsies. Diagnosis can be challenging, especially for small, well-differentiated cancers. Recently, machine learning algorithms have been developed for detecting PrCa in whole slide images (WSIs) with high test accuracy. However, the impact of these artificial intelligence systems on pathologic diagnosis is not known. To address this, we investigated how pathologists interact with Paige Prostate Alpha, a state-of-the-art PrCa detection system, in WSIs of prostate needle core biopsies stained with hematoxylin and eosin. Three AP-board certified pathologists assessed 304 anonymized prostate needle core biopsy WSIs in 8 hours. The pathologists classified each WSI as benign or cancerous. After ~4 weeks, pathologists were tasked with re-reviewing each WSI with the aid of Paige Prostate Alpha. For each WSI, Paige Prostate Alpha was used to perform cancer detection and, for WSIs where cancer was detected, the system marked the area where cancer was detected with the highest probability. The original diagnosis for each slide was rendered by genitourinary pathologists and incorporated any ancillary studies requested during the original diagnostic assessment. Against this ground truth, the pathologists and Paige Prostate Alpha were measured. Without Paige Prostate Alpha, pathologists had an average sensitivity of 74% and an average specificity of 97%. With Paige Prostate Alpha, the average sensitivity for pathologists significantly increased to 90% with no statistically significant change in specificity. With Paige Prostate Alpha, pathologists more often correctly classified smaller, lower grade tumors, and spent less time analyzing each WSI. Future studies will investigate if similar benefit is yielded when such a system is used to detect other forms of cancer in a setting that more closely emulates real practice.

Histopathologic False-positive Diagnoses of Prostate Cancer in the Age of Immunohistochemistry

There are few studies into the rate and causes of histopathologic false-positive diagnosis of prostate cancer. Only 2 of these, including a previous one from our group, incorporate survival data. In addition, in none of the previous studies had immunohistochemistry (IHC) been originally requested on any of the misdiagnosed cases. Diagnostic biopsies (n=1080) and transurethral resection of prostate specimens (n=314) from 1394 men with clinically localized prostate cancer diagnosed in the United Kingdom but treated conservatively between 1990 and 2003 were reviewed by a panel of 3 genitourinary pathologists. Thirty-five cases were excluded for being potentially incomplete. Of the remaining 1359, 30 (2.2%) were reassigned to a nonmalignant category (26 benign and 4 suspicious for malignancy). IHC had been originally performed on 7 of these. The reasons for the errors were recorded on each case: adenosis (19), partial atrophy (3), prostatic intraepithelial neoplasia (2), seminal vesicle epithelium (1), and hyperplasia (1). Follow-up of these men revealed only one prostate cancer-related death, possibly due to unsampled tumor. In conclusion, a relatively small number of prostate cancer mimics were responsible for a large proportion of the false-positive prostate cancer diagnoses and the use of IHC did not prevent the overcall of benign entities as cancer in approximately a quarter of these cases. Targeting these mimics at educational events and raising awareness of the pitfalls in the interpretation of IHC in prostate cancer diagnosis, emphasizing that glands within a suspicious focus should be treated as a whole rather than individually, may be beneficial in lowering the rate of false-positive diagnosis.

False-negative prostate needle biopsies: frequency, histopathologic features, and follow-up

Little is known about the frequency, histopathologic characteristics, and clinical consequences of false-negative prostate biopsies, that is, biopsies classified as benign but containing adenocarcinoma or atypical suspicious glands [atypical small acinar proliferations (ASAP)]. Objective of this study was to evaluate false-negative prostate biopsy in a prostate cancer screening setting. Prostate biopsy sets of 196 participants of a screening trial, which had been reported as "benign" at initial diagnosis, followed by a diagnosis of adenocarcinoma in a subsequent screening round were reviewed by 2 urologic pathologists. Adenocarcinoma was identified in 19 biopsy cores corresponding to 16 (8.2%) patients and ASAP in 24 cores, corresponding to 19 patients (9.7%). All missed prostate cancers were Gleason score 6 (3+3). After correction for patient selection, the overall false-negative biopsy rate was estimated to be 2.4%; 1.1% for prostate cancer; and 1.3% for ASAP. Clinicopathologic features at the time of initial biopsy and of subsequent prostate cancer diagnosis did not differ between patients with a false-negative or true benign biopsy. Relatively low number of atypical glands (<10 glands), intense intermingling with preexistent glands or lack of architectural disorganization were the most prominent risk factors for a false-negative diagnosis. Another potential pitfall was the presence of prostate cancer variants, as 1 adenocarcinoma was of foamy gland type and 3 of pseudohyperplastic type. Routine examination of at least 1 level of prostate biopsy sets at high magnification and awareness of histologic prostate cancer variants might reduce the risk of missing or misinterpreting a relevant lesion at prostate biopsy evaluation.

Diagnostic value of systematic biopsy methods in the investigation of prostate cancer: a systematic review

PURPOSE

Several new extended prostate biopsy schemes (greater than 6 cores) have been proposed. We compared the cancer detection rates and complications of different extended prostate biopsy schemes for diagnostic evaluation in men scheduled for biopsy to identify the optimal scheme.

MATERIALS AND METHODS

In a systematic review we searched 13 electronic databases, screened relevant urological journals and the reference lists of included studies, and contacted experts. We included studies that compared different systematic prostate biopsy methods using sequential sampling or a randomized design in men scheduled for biopsy due to suspected prostate cancer. We pooled data using a random effects model when appropriate.

RESULTS

We analyzed 87 studies with a total of 20,698 patients. We pooled data from 68 studies comparing a total of 94 extended schemes with the standard sextant scheme. An increasing number of cores were significantly associated with the cancer yield. Laterally directed cores increased the yield significantly (p = 0.003), whereas centrally directed cores did not. Schemes with 12 cores that took additional laterally directed cores detected 31% more cancers (95% CI 25 to 37) than the sextant scheme. Schemes with 18 to 24 cores did not detect significantly more cancers. Adverse events for schemes up to 12 cores were similar to those for the sextant pattern. Adverse event reporting was poor for schemes with 18 to 24 cores.

CONCLUSIONS

Prostate biopsy schemes consisting of 12 cores that add laterally directed cores to the standard sextant scheme strike the balance between the cancer detection rate and adverse events. Taking more than 12 cores added no significant benefit.

Virtual microscopy in prostate histopathology: simultaneous viewing of biopsies stained sequentially with hematoxylin and eosin, and alpha-methylacyl-coenzyme A racemase/p63 immunohistochemistry

PURPOSE

Histopathological diagnosis of small focus carcinomas in prostatic needle biopsies is often assisted by IHC. To make a definitive diagnosis the pathologist must compare IHC findings with hematoxylin and eosin stained tissue morphology. We introduce what is to our knowledge a new application of virtual microscopy, in which hematoxylin and eosin, and IHC stains done sequentially on the same microscope slide can be simultaneously displayed and compared on a computer screen.

MATERIALS AND METHODS

A total of 30 hematoxylin and eosin stained prostatic needle biopsies were scanned with a computer controlled microscope. The slides were destained and then immunostained with a cocktail of AMACR and p63 antibodies, which labels the nuclei of nonmalignant basal cells (p63) and the cytoplasm of neoplastic glandular cells suspicious for malignancy (AMACR). The slides were then scanned again and the pairs of virtual slides were aligned for synchronized viewing.

RESULTS

The presented technique was found helpful when suspicious lesions were small and when examining the immunoprofile of specimens was warranted, in addition to examining hematoxylin and eosin stained tissue morphology. The usefulness of our approach based on virtual microscopy can be evaluated on the website , which also serves as an educational tool for self-learning the correlation between hematoxylin and eosin stained tissue morphology, and AMACR/p63 IHC in prostate biopsies.

CONCLUSIONS

The technology for simultaneously viewing sequentially hematoxylin and eosin and IHC stained prostate biopsies can be readily used for educational purposes, as exemplified by our website, and along with the availability of rapid virtual slide scanners it can also be used for clinical diagnostics.