|
 
 |
| ORIGINAL ARTICLE |
|
| Year : 2023 | Volume
: 13
| Issue : 1 | Page : 139-143 |
|
Audit of prostate cancer diagnosis using P63 and prostate-specific antigen immunostains at a tertiary care hospital in South-East Nigeria
Darlington C Amadi1, Chinedu Onwuka Ndukwe2, Christopher C Obiorah3, Cornelius Ozobia Ukah2
1 Department of Histopathology, Federal Medical Centre, Owerri, Nigeria
2 Department of Anatomic Pathology and Forensic Medicine, Nnamdi Azikiwe University, Nnewi Campus, Nnewia, Nigeria
3 Department of Anatomic Pathology, University of Port Harcourt Teaching Hospital, Port Harcourt, Nigeria
| Date of Submission | 20-Jul-2022 |
| Date of Acceptance | 27-Sep-2022 |
| Date of Web Publication | 06-Dec-2022 |
Correspondence Address:
Dr. Chinedu Onwuka Ndukwe
Department of Anatomic Pathology and Forensic Medicine, Nnamdi Azikiwe University, Nnewi Campus, Nnewi
Nigeria
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/aihb.aihb_139_22
Introduction: Prostate cancer (CaP) is a leading cancer diagnosis and cause of cancer-related deaths among Nigerian men, and the accuracy of the pathologic diagnosis is critical for optimal patient care. This study aims to define the relative proportional accuracy of the histological diagnosis of prostatic cancer with hematoxylin and eosin (H and E) slides at Federal Medical Centre (FMC) Owerri, Nigeria. It also seeks to determine Gleason's grading and grade group pattern of CaP in our hospital. Materials and Methods: This study is a 5-year retrospective study reviewing all the H and E slides of prostatic Tru-Cut biopsies that were already diagnosed as cancer at FMC Owerri. These diagnoses were read and confirmed using their morphological appearance on light microscopy. Those that were suspicious, especially those that were diagnosed with well-differentiated adenocarcinoma (Gleason 3 + 3) and poorly differentiated carcinoma (Gleason 5 + 5), were sent for immunohistochemical studies. The well-differentiated carcinomas were stained for basal cells using P63, whereas poorly differentiated carcinomas were stained with prostate-specific antigens to confirm tissue of origin. These are to differentiate them from benign mimickers and metastatic carcinomas, respectively. Results: The age range of the patients was between 50 and 99 years, with a mean age of 72.84 years and peak age in the 70–79 years of age group. Using immunohistochemistry on the specimens which fitted into the criteria for immunohistochemistry, this study showed the relative proportional accuracy of the histological diagnosis of prostatic cancer using H and E to be 95.55% and the positive predictive value of H and E to be 0.83. With respect to International Society of Urological Pathology (ISUP) group grading, high-grade cancers predominated, and the most predominant grade is 5. Conclusion: H and E stain has high diagnostic accuracy (95.55%), but a 4.45% diagnostic error using H and E is quite significant. Immunohistochemistry, therefore, should act as a most valuable adjunctive, which significantly increases the diagnostic accuracy of prostatic carcinoma diagnosis.
Keywords: Gleason grading, haematoxylin and eosin, immunohistochemistry, Nigeria, prostate cancer
How to cite this article:
Amadi DC, Ndukwe CO, Obiorah CC, Ukah CO. Audit of prostate cancer diagnosis using P63 and prostate-specific antigen immunostains at a tertiary care hospital in South-East Nigeria. Adv Hum Biol 2023;13:139-43 |
How to cite this URL:
Amadi DC, Ndukwe CO, Obiorah CC, Ukah CO. Audit of prostate cancer diagnosis using P63 and prostate-specific antigen immunostains at a tertiary care hospital in South-East Nigeria. Adv Hum Biol [serial online] 2023 [cited 2023 Mar 27];13:139-43. Available from: https://www.aihbonline.com/text.asp?2023/13/1/139/362699 |
| Introduction | |  |
Prostate cancer (CaP) is the most common heterogeneous, extracutaneous and internal malignancy amongst men in Nigeria. Adenocarcinoma is the most common form.[1],[2],[3],[4] This cancer is a disease of increasing significance worldwide, and Nigeria is classified in the low-incidence CaP group, although with high mortality, there is no countrywide population-based cancer registration to enable a proper assessment.[5],[6],[7]
The American Cancer Society enumerated the risk factors of CaP to include men of old age (50 years and above), African ancestry and men living in geographical locations such as North America, Northwestern Europe, Australia and the Caribbean islands. Other risk factors are: positive family history, inherited genetic changes such as inherited mutations of the BRCA1 or BRCA2 genes and men with Lynch syndrome. Lifestyles such as a high saturated fat diet, obesity and smoking are also risk factors.[2]
The prostate gland consists microscopically of glands lined by double layers of cells: inner secretory cells and the outer basal cell.[2],[4] Positive stains for basal cells are CK903 (34 βE12/high-molecular-weight keratin), p63 and CK5/6. In addition, basal cells express androgen receptors.[3] The presence of basal cells differentiates prostatic atrophy and other benign mimickers (basal cells present) from adenocarcinoma (basal cells not present).[3] The inner secretory cells stain positive for prostate-specific antigen (PSA) immunostain.
The accuracy of the pathologic diagnosis of this malignant tumour is critical for optimal patient care.[1] The histologic diagnosis of CaP is based mainly on its morphology and architecture, as seen in the light microscope. The accurate diagnosis and grading of this tumour, based on examination of its glandular architecture on light microscopy at low-to-medium power in haematoxylin and eosin (H and E)-stained sections, has been proposed by Gleason.[8]
This Gleason system of grading has, however, evolved over the years with changes made to the original Gleason grading and scoring at different ISUP consensus meetings and by the WHO. This continuous evolution and the limitations of interobserver differences in diagnosis, coupled with the difficulty of differentiating metastasis from poorly differentiated (5 + 5) CaP and that of ruling out benign mimickers from Gleason score 3 + 3 CaP solely on H and E, has led to this study.
Regular audit of the previous diagnosis is essential in the improved management of this disease. This has therefore made it important to re-evaluate the practice in our department using current guidelines as a form of audit since accurate diagnosis is key to the prognostication and good management of the disease. This will help to establish a good management protocol in this centre.
This study aims to audit all the CaP diagnoses made on Tru-Cut biopsies at Federal Medical Centre (FMC), Owerri, during the study to determine the accuracy of the histologic diagnosis using basal cell immunohistochemical marker (p63) for the suspicious cases diagnosed as well-differentiated adenocarcinoma (Gleason score 3 + 3) and tissue PSA immunostain for those diagnosed as poorly differentiated adenocarcinoma (Gleason score 5 + 5). This is to rule out benign mimickers and metastatic carcinoma to the prostate, respectively. The study also aims to determine Gleason's grading, grade group patterns and age demographics of CaP seen at FMC, Owerri.
| Materials and Methods | | |
Materials
This is a 5-year retrospective study of all Tru-Cut prostate biopsy specimens diagnosed as carcinoma at FMC, Owerri, between 1st January 2016 and 31st December 2020. Materials for this study consist of histological slides and tissue blocks of Tru-Cut biopsy specimens which are already diagnosed as prostatic adenocarcinoma in the histopathology laboratory of FMC, Owerri.
Inclusion criteria
Prostatic Tru-Cut biopsies already diagnosed as prostatic carcinoma with complete demographic and clinical records in the request or report forms were included in the study, provided that their slides were well preserved and those with broken slides have properly preserved blocks of paraffin-embedded tissues.
Exclusion criteria
Suspicious cases whose paraffin-embedded tissue blocks were unavailable or insufficient were excluded from the study.
Methods
Relevant biodata and clinical information of the cases were retrieved from records in the surgical day book, laboratory request forms and/or histology reports book of the department. These biodata include age, clinical features and other relevant ancillary investigations. The H- and E-stained slides of the selected cases were retrieved from the departmental archives, and where necessary, reproduced and then reviewed.
Review of diagnosis
During the review, the diagnoses were read and confirmed using their morphological appearance on light microscopy and guided by the ISUP 2014 group grading system, which categorised CaP from Gleason 3 + 3. Those that were suspicious, especially those that were diagnosed with well-differentiated adenocarcinoma (Gleason 3 + 3) and poorly differentiated carcinoma (Gleason 5 + 5), were sorted from all other cases. These suspicious cases were then subjected to immunohistochemical analysis.
Processing of suspicious cases
The paraffin-embedded blocks were processed using immunostains for basal cells (P63) and tissue of origin (PSA). P63 was used for well-differentiated adenocarcinoma (Gleason 3 + 3) and PSA for cases diagnosed initially as poorly differentiated (Gleason 5 + 5) carcinoma; this is to differentiate them from perceived benign mimickers and metastasis from nearby tissues, respectively.
Ethical clearance
Ethical clearance for the study was obtained from the FMC, Owerri Research Ethical Committee.
Data analysis
The data were analysed using SPSS Statistics for Windows version 21.0, IBM Corp. Armonk, NY, USA.
| Results | | |
A total of 108 prostate biopsies (all Tru-Cut biopsy samples) were examined. After review, only 96 cases met the inclusion criteria and hence were included in the study. Out of the 96 samples, 93 were 55 years and above, whereas only three were below 55 years, but all were above 50 years. The mean age was 72.84 years. Its frequency increases from 50 years, peaks between 70 and 79 years and then starts to wane [Table 1].
Using the Gleason score scheme, our study shows that high-grade cancers predominated, and the most frequent single score is 9 [Table 2]. The predominant grade pattern using ISUP grouping is shown to be of the ISUP grade group 5 (high grade) [Figure 1]. | Table 2: Frequency of the various Gleason scores before and after review
Click here to view |
Twenty-nine cases out of the 96 were sent for immunohistochemical analysis. Twenty-three (79.31%) of these samples sent for immunohistochemistry were well differentiated (Gleason 3 + 3) and were evaluated using P63 immunostain, whereas six poorly differentiated cases (Gleason 5 + 5) were sent for evaluation using PSA immunostain. Of the 29 cases subjected to immunohistochemistry, six could not be evaluated, bringing our total sample down to 90. Out of the 90 samples, four samples were not confirmed as cancer after immunohistochemistry [Table 3] and [Figure 2]. Therefore, the accuracy of H and E stain without adjunctive study is 95.6% (86/90 cases). The positive predictive value (PPV) of diagnosing CaP using routine H and E at FMC, Owerri, is 83%. | Table 3: Immunohistochemistry results for p63 and prostate-specific antigen
Click here to view |
 | Figure 2: p63-positive benign prostatic glands (a), p63-negative malignant prostatic glands (b) and PSA-positive poorly differentiated prostatic carcinoma (c). PSA: Prostate-specific antigen.
Click here to view |
| Discussion | | |
In this study, we sought to define the relative proportional accuracy of the histological diagnosis of prostatic cancer and to determine the Gleason's grading, grade group pattern and age demographics of CaP in our hospital. We found the relative proportional accuracy of the histological diagnosis of prostatic cancer using H and E to be 95.55% and the PPV of H and E to be 0.83. Using ISUP group grading, high-grade cancers predominated, and the most predominant grade was 5. The modal age group for CaP in our environment is 70–79 years.
There are a few limitations to our study. One of the limitations is that our study considered only cancer cases that were diagnosed using Tru-cut biopsy samples. This is because in this centre, mainly Tru-cut biopsies are presented to the histopathology department. Other encountered limitations were that some of the tissue blocks sent for immunohistochemistry were inadequate for analysis and thus were not evaluated, whereas some of the H and E slides that were missing were not reproduced because the blocks were poorly preserved and hence diminishing the sample size.
This study observed a 4.45% change in diagnosis from prostatic adenocarcinoma to benign prostatic lesion, with the diagnostic accuracy of H and E being 95.55%. This diagnostic error using H and E is quite significant considering the increasing samples of prostate tissue in the workload of pathologists in the laboratory and the high percentage of male patients expected to be diagnosed with CaP in their later lives (above 50 years).
Kumaresan et al.[9] observed overall diagnostic change in 17 of 1034 cases (1.64%) of prostatic disease in their study. However, only one sample representing 0.25% changed from malignant to benign in the 399 malignant cases they reviewed. This percentage error in diagnosis is quite low compared to our study, which recorded a 4.45% diagnostic change from cancer to benign.
Anand et al.[10] observed in their study that even though the diagnosis of CaP is mainly based on architectural patterns and cytology, immunohistochemistry also acts as a valuable adjunctive, which significantly increases the diagnostic accuracy in prostatic carcinoma. They observed a 50% disparity between when H and E are used alone and when immunohistochemistry is used as an adjunctive in diagnosis. This result is significantly different from the result of our study, which also is quite different from that of Kumaresan et al.[9] but can be explained by the fact that our study included only cases diagnosed as cancer with H and E stain. The disparity between Kumaresan et al.[9] and Anand et al.[10] cannot be explained, and the high percentage margin between our study and Kumaresan et al.[9] when they reviewed the 399 cases initially diagnosed as cancer can also not be explained.
Using the Gleason score scheme, our study shows that high-grade cancers predominated, and the most frequent single score is 9. Similarly, Obiorah and Nwosu[11] in their study observed the same pattern with the most frequent single score of 8. Bassey et al.,[12] a study in Calabar, recorded lower predominant Gleason scores of 6 and 7, differing from our results in Owerri, Obiorah and Nwosu.[11] There is no available explanation for this and could probably be due to the lack of patients' interest and the absence of screening programs in these two cities (Owerri and Port Harcourt) as observed by Oranusi et al.[13] in Anambra, South-East Nigeria. Calabar being more metropolitan than Owerri, it is probable that patients may have been presenting earlier in Calabar, but that of Port Harcourt was not explained by Obiorah and Nwosu.[11] Most patients do not have early symptoms, which would have prompted them to seek early medical treatment, and so the early grades and scores were missed in Port Harcourt and Owerri.
Using the ISUP group grading system, our study observed that high-grade cancers predominated, and the most predominant grade was group 5 with 41.5%. The authors found this to be in agreement with the only study from Southeast Nigeria reporting on ISUP grade group patterns.[14] Our finding of the predominance of high-grade (grade group 5) CaP in our centre also agrees with another study from Southwest Nigeria.[15]
Furthermore, in this study, it was observed that using Gleason's pattern of scoring, interobserver variation in grading and scoring persisted. Similarly, Mcleam et al.[16] noticed this pitfall of interobserver variation, and therefore, emphasised the importance of a central review process for pathology grading in clinical trials, especially where the treatment is directly affected by this grading and scoring information.
In this study, the age range of the patients was between 50 and 99 years, with a mean age of 72.84 years and peak age in the 70–79 years of age group. Our finding compares favourably with a similar study by Obiorah and Nwosu[11] in Port Harcourt, South-South Nigeria, where the age range was between 42 and 90 years, with the mean at 70 years and peak at 70–79 years. Furthermore, a study in Benin City, South-South Nigeria, by Forae and Aligbe[17] showed a peak age range of 70–79 years and a mean age of 68 years. However, Akang et al.,[18] in a similar study in Benin City, South-South Nigeria, observed a peak period in the seventh decade, a decade lower than what is found in Owerri and Port Harcourt and from a similar study by Forae and Aligbe[17] in the same city. These other studies (our study, Obiorah and Nwosu[11] and Forae et al.)[17] are more recent (2018, 2011 and 2014, respectively) than that by Akang et al.[18] which was conducted in 1996, more than a decade earlier.
| Conclusion | | |
Even though using H and E stain has high diagnostic accuracy (95.55%) and is the bedrock of CaP diagnosis, which is based mainly on the architectural pattern, cytological and ancillary features, a 4.45% diagnostic error is quite significant. Immunohistochemistry, therefore, acts as a more valuable adjunctive, which significantly increases the diagnostic accuracy of suspicious cases in prostatic carcinoma diagnosis. CaP in this environment is predominantly in the high ISUP grade group.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | |
| 2. | |
| 3. | Epstein IJ. The lower urinary tract and male genital tract system. In: Kumar V, Abbas AK, Fuasto N, Aster JC, editors. Robins and Cortran Pathologic Basis of Diseases. 8 th ed. Philadelphia: Elsevier; 2010. p. 971-1004.  |
| 4. | Mohammed AZ, Edino ST, Ochicha O, Gwarzo AK, Samaila AA. Cancer in Nigeria: A 10-year analysis of the Kano cancer registry. Niger J Med 2008;17:280-4. |
| 5. | Ifere GO, Abebe F, Ananaba GA. Emergent trends in the reported incidence of prostate cancer in Nigeria. Clin Epidemiol 2012;4:19-32. |
| 6. | Ndukwe CO, Ndukwe CC, Ike KG, Eziagu UB. Cancer mortality pattern in a resource-poor country: A case study of a teaching hospital in the South-East region of Nigeria. Ibnosina J Med Biomed Sci 2021;13:67-73. [Full text] |
| 7. | Akpuaka SO, Clarke-Tasker VA, Nichols-English GJ, Daniel MG, Akpuaka AI. Knowledge and perceptions of prostate cancer among Nigerian male immigrants. ABNF J 2013;24:23-7. |
| 8. | Humphrey PA. Gleason grading and prognostic factors in carcinoma of the prostate. Mod Pathol 2004;17:292-306. |
| 9. | Kumaresan K, Kakkar N, Verma A, Mandal AK, SinghSK, Joshi K et al. Diagnostic utility of α-methylacyl CoA racemase (P504S) and HMWCK in morphologically difficult prostate cancer. Diagn Pathol 2010;5:83. |
| 10. | Anand OD, Rao PS, Kishore Kumar C, Vijaya Bhaskar RV. Diagnostic utility of immunohistochemical markers in prostate cancer. IOSR J Dent Med Sci 2017;3:22-8. |
| 11. | Obiorah CC, Nwosu SO. A histopathological study of carcinoma of the prostate in Port Harcourt, Nigeria. Niger J Clin Pract 2011;14:363-7. [ PUBMED] [Full text] |
| 12. | Bassey EI, Edoise MI, Ayi D. Prognostication studies of prostate cancer in black Africa: Findings from Calabar, South-South, Nigeria. Int J Contemp Med Res 2018:E1-5. |
| 13. | Oranusi CK, Mbieri TU, Oranusi I, Nwofor AM. Prostate cancer awareness and screening among male public servants in Anambra State, Nigeria. Afr J Urol 2012;18:72-4. |
| 14. | Madubuike KC, Ndukwe CO, Chiemeka ME, Ezejiofor IF, Ogbu CC, Onyiaorah IV, et al. Audit of prostatic diseases diagnosed from Tru-Cut biopsies in Nnamdi Azikiwe University Teaching Hospital, Nnewi, using the P63 immunohistochemical marker. Hamdan Med J 2022;15:78-82. [Full text] |
| 15. | Omenai SA, Mashor IM, Chibuzo IN, Adebayo S, Shittu OB, Ogunbiyi JO. Histomorphological characteristics and distribution of prostatic carcinoma foci using multiple site-specific labelled core biopsies in a tertiary hospital in southwestern Nigeria. JCTI 2021;11:18-27. |
| 16. | McLean M, Srigley J, Banerjee D, Warde P, Hao Y. Interobserver variation in prostate cancer Gleason scoring: Are there implications for the design of clinical trials and treatment strategies? Clin Oncol (R Coll Radiol) 1997;9:222-5. |
| 17. | Forae GD, Aligbe JU. Histopathological patterns of prostate cancer in an African population: A private practice experience. Trop J Med Res 2014;17:16-9. [Full text] |
| 18. | Akang EE, Aligbe JU, Olisa EG. Prostatic tumours in Benin City, Nigeria. West Afr J Med 1996;15:56-60. |
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]
|
Search Pubmed for