Evolution of Androgenic Deprivation in Treatment of Prostate Cancer in Kinshasa ()
1. Introduction
Prostate cancer (PCa) is the most common cancer in men in America (Afro-Americans and Afro-Caribbean) and Europe [1] [2] [3]. Patients with localized or sometimes locally advanced forms benefit from radical prostatectomy or radiotherapy. On the other hand, in patients with aggressive or metastasized forms; the treatment options may be radiotherapy but more commonly androgen deprivation [4]. To reduce the plasma level of testosterone and its derivative, dihydrotestosterone (DHT), androgen deprivation; apart from surgical castration still used by Anglo-Saxons, other practitioners prefer hormone therapy [5]. This hormone therapy has evolved with the emergence of LH-RH analogs, then treatment with non-steroidal anti-androgens, and finally, more recently, LH-RH antagonists, and Gn-RH agonists [6]. Depending on the patient’s response, this hormone therapy goes from the first to the third line [7] [8]. Over 90% of patients treated respond to androgen deprivation. Unfortunately, transiently, its duration varies from a few months to a few years. The median being 12 to 18 months, then relapse is observed in 100% of cases. After the escape from this first line of hormone therapy, responses to alternative hormonal manipulations are rare, the tumor is resistant to castration [7] [8]. Once the hormone-independence stage has been reached, the tumor is resistant to castration. Median overall survival is 34 months [6]. Castration-resistant PC (CRPC) is an advanced form characterized by disease progression after surgical or pharmaceutical castration (androgen deprivation). The process by which prostate cancer cells become resistant to castration is not clear, but androgenic privations have been shown to offer a selective advantage to androgen-independent cells, which eventually grow and repopulate the tumor [9]. Compared with castration sensitive PCa, the prognosis of patients with CRPC is poor and survival is reduced. Until very recently, treatment options were mainly limited to symptomatic relief of bone metastases, which are more common in CRPC than in the castration-sensitive form [9] [10] [11] [12]. To provide a clear picture of the burden of CRPC, one must consider the prevalence of the disease, the relative time of onset versus diagnosis, patient characteristics including demographics, comorbidity, the onset of disease, metastatic form, and probable survival. There is, however, insufficiency of epidemiological evidence specifically characterizing CRPC outside of the settings of controlled trials in which patients may not represent the general population and normal disease progression. This can lead to its sub-optimal management; for example, the identification of patients with CRPC who are at risk of developing metastases is currently hampered by a poor understanding of its real epidemiology. Identifying people with CRPC may seem straightforward after androgen deprivation (drug or surgical). The characterization of the disease in epidemiological terms, eg incidence, prevalence, and survival, is however less clear. This can be attributed at least in part to the difficulty of defining, and therefore of studying, the patient population. The varying terminology—CRPC, HRPC (Hormone-refractory PCa), AIPC (Androgen-independent PCa), ERPC (Endocrine-resistant PCa)—reflects subtle differences in dentition that may hamper research comparison. Practitioners can also use a variety of diagnostic methods: prostate-specific antigen (PSA) assays, the development of metastases, or other factors to determine if a patient is defined as CRPC. The recently published European Association of Urology (EAU) guidelines aim to standardize the diagnosis of CRPC and include a list of five defining factors of CRPC [4].
These are as follows:
● Serum testosterone level.
● Three consecutive increases in Prostate Specific Antigen (PSA) 2 weeks apart resulting in two increases of 50% over the nadir.
● Anti-androgen stops for at least 4 weeks.
● Progression of PSA despite secondary hormonal manipulations.
● Progression of bone or soft tissue damage.
CRPC is a heterogeneous disease, and despite the availability of such guides for diagnosing CRPC, in practice, this can vary. Also, besides, the routes of treatment and clinical practice, particularly the stage of the disease at the onset of androgen deprivation therapy, vary widely between geographic locations and even the individual’s clinic. Therefore, establishing common epidemiological estimates for the CRPC population becomes very complex and may become less relevant for individual scenarios [13].
In our environment, most affected patients consult at the advanced stage of the disease; thus, justifying androgen deprivation as a mode of treatment.
This study aimed to improve the clarity of the epidemiological evidence around CRPC, by identifying, assessing, and describing the most relevant elements that characterize the affected patient population using observational data.
Our objective was to assess the responses to hormonal deprivation, patient survival and to identify the different predictors of mortality.
2. Methods
Nature, Period, Framework and parameters of interest.
This is a retrospective observational epidemiological study that focused on the evolution of the 51 patients followed for PCa at the Pointe à Pitre Clinic during the period from March 2014 to June 2018. Pointe-à-Pitre Clinic is a non-profit organization, non-denominational and nonpolitical called “PROSTATE CANCER VIGILANCE AFRIQUE CENTRALE’’, PCVAC in acronym was created in Kinshasa, capital of the DRC on April 27, 2016. The head office of the association is in Kinshasa, within the Clinic cited above in the city of Kinshasa province in Matete Township, Kinzazi district, n˚ 11D in Matete Health District. During the entire study period, 1364 patients were received at the CPAP and we identified 165 cases of prostate cancer (12%) and among them, only 51 cases were retained to constitute our sample of coverage. Among the 165 patients, many were excluded because their follow-up was incomplete (PSA and Testerone). Our parameters of interest were age, profession, marital status, place of residence, complaints, cTNM, PSA, testosterone biopsy, prostate ultrasound, MRI results, Scintigraphy, radiography (pelvis and column), histological result, Gleason score, stage of progression, treatment, patient survival, a predictor of mortality. The androgen deprivation method has been either surgical (bilateral orchidectomy) or chemical (hormone therapy). We used for hormonotherapy, Cyproterone acetate (50 - 100 mg), Gosereline (10.8 mg) Bicalutamide (50 mg), and Docetaxel for the chemotherapy. Some patients who have undergone radical prostatectomy have received complete hormone therapy or orchidectomy for PCa recurrence.
2.1. Inclusion Criteria
We considered patients’ files with a PCa which was treated and operated during the period of our study.
2.2. Non Inclusion Criteria
Incomplete or absent files on during data collecting have not been taken into account.
2.3. Collection of Data
We collected data by completing an ad hoc form related to the documentary review focused on the medical records and registers of patients.
3. Statistical Analyses
The data were computerized using Excel 2010 software and were analyzed using SPSS version 17 software. Tables or graphs were used, as appropriate, for the presentation of the results. The continuous quantitative variables with Gaussian distribution were presented as mean ± standard deviation; those with non-normal distribution in the form of the median (extremes). Qualitative variables were described as relative frequency (%). Comparison of proportions, medians, and means was performed using Chi-square, Mann Whitney Wilcoxon, and Student’s t-tests, respectively. Independent determinants of resistance to surgical castration and PSA ≥ 100 ng/ml were identified using logistic regression. Kaplan Meier’s method estimated the probability curve of resistance to surgical castration. It also described survival between the date of diagnosis of CaP and death (complete data) and the end of the study (censored data). The Log-rank test was used to compare survival curves. Cox’s regression looked for independent predictors of mortality.
A p-value < 0.05 was considered the threshold of statistical significance.
Ethical Considerations
During the collection and analysis of our data, confidentiality was strictly enforced.
4. Results
4.1 General Characteristics of the Study Population
About 51 patients, the average age was 69.4 ± 9.7 years, with extremes of 40 to 92 years. The age range of 70 to 79 years was the most common (39.2%). Obstructive signs and irritative signs were respectively 72.5% and 33.3%. Dysuria was the obstructive symptom that most prompted patients to consult (45.1%) followed by nycturia (23.5%) like irritative symptoms. Hypertension was the most common comorbidity (25.5%). It appears that; 47.1% of PCa were diagnosed at the metastasis stage, followed by cancers at high risk of progression (25.5%). Only 7.8% of cancers were at low risk. The metastases were from localizations variables; bone (50%), multiples (20%), ganglion (16.7%) and testicles (12.5%). The year 2014 saw more consultations (37.2) and the highest mortality rate was observed during the year 2017 (50%). The overall mortality rate was 19.6% and overall survival was 80.4% (Table 1).
Table 1. General characteristics of the population of the study.
*Global mortality in 4 years.
4.2. Treatment and Evolution
The treatment varied depending on the case (Hormone therapy, Surgery, chemotherapy):
- Ciproterone Acetate (Androcure) 50.9%;
- The Goserelin-Bicalutamide combination in 45.0%;
- Surgical castration in 23.5%;
- Surgical castration was associated with TURP in 11.8% of cases;
- Radical prostatectomy in 3.9%.
From an evolutionary point of view; the rate of castration resistance was 43.1% within a median of 1.4 (1 - 3) years of response to treatment (Table 2).
4.3. Evaluation of Castration Resistance
Castration resistance was observed from the 5th month of treatment, especially for carrier patients with metastases.
4.4. Resistance to Castration According to Gleason Score
Patients with a Gleason score between 8 and 10 had a higher frequency of resistance compared to those with a Gleason score between 6 and 7; log-rank test (p = 0.018) (Figure 1).
4.5. Risk of Resistance According to the D’AMICO Classification
According to D’AMICO’s classification, the risk of castration resistance was variable:
- Twice for the intermediate-risk PC [OR 2.02 95% CI (1.45 - 3.90); P = 0.021];
- Three for high-risk PC [OR 2.95 95% CI (1.36 - 4.69); P = 0.041];
- And 6 times for metastatic CaP [OR 5.88 95% CI (1.62 - 7.99); P = 0.019] (Table 3).
Table 2. Distribution of patients according to treatment mode.
Table 3. Distribution of resistance according to the D’AMICO classification.
Figure 1. Gleason score and castration resistance.
4.6. Resistance to Castration According to Clinical Signs
Dysuria came first in 39.1%, followed by pollakiuria in 30.4%, nocturia in 26.1%, and bone pain in 21.1% of cases (Table 4).
4.7. Assessment of Prostate Specific Antigen Rate and Testosterone (ng/ml)
During treatment, the PSA level tended to decrease for all patients. Its increase has been observed in some patients from the 4th dosage. The mean PSA was 51.5 (0.3 - 2528.7) ng/ml for un median of 7.75 ng/ml. However, the testosterone level which reached the castration rate still tended to cancel out. Its average was 2.8 ng/ml (0.5 - 8.15).
4.8. Prostate Specifin Antigen Evaluation According to Age, cTNM, and Gleason Score
The PSA level was not statistically significant (p 0.779) compared to the ages of the patients. The increase in PSA level was influenced by clinical stage cT3 - cT4 [OR 15.0 95% CI (2.02 - 17.11); p = 0.006], with a statistically significant difference in cT1 - cT2 (p 0.006) and score Gleason [OR 6.07 95% CI (1.49 - 24.76)] without any statistically significant difference in score 6 - 7 (p = 0.011) (Table 5).
Table 4. Distribution of clinical signs in patients resistant to castration.
Table 5. Distribution of Prostate Specific Antigen levels by age, cTNM, and Gleason score.
4.9. Assessment of Patients’ Survival
The probability of patient survival was 92.2% at 10 months, 84.3% at 15 months, 82.4% at 20 months, and 80.4% at 48 months, respectively. The median patient survival was 30 (24 - 30) months and the mean survival was 26.6 months.
4.10. Assessment of Patients’ Survival versus Castration Resistance
Patients resistant to castration had significantly reduced survival compared to those who did not (p = 0.029) (Figure 2).
4.11. Assessment of Survival about Prostate Specific Antigen Level
The survival of patients with a PSA level ≥ of 100 ng/ml (p = 0.006) was significantly lower compared to the others (Figure 3).
Figure 2. Patients’ survival versus castration resistance.
Figure 3. Survival curve as a function of the prostate specific antigen level.
4.12. Assessment of Survival about Gleason Score
The survival of patients with a Gleason score of 8 - 10 (p 0.004) was significantly lower than those with a score of 6 - 7 (Figure 4).
4.13. Predictors of Mortality
In univariate analysis; PSA levels ≥ 100 ng/ml [HR 10.20 95% CI (1.29 - 13.56); p = 0.001], Gleason score 8 - 10 [HR 10.97 98% CI (1.39 - 16.68); p = 0.035] castration resistance [HR 3.98 95% CI (1.56 - 7.04); p = 0.017] and metastases [HR 2.67 95% CI (1.69 - 10.35), p = 0.007] were predictors of mortality, without any
significant difference within each group (Table 6). In multivariate analysis, the Gleason score 8 - 10 [HRa 10.15 95% CI (2.18 - 12.23); p = 0.035] and the PSA ≥ 100 [HRa 8.49 95% CI (2.15 - 10.56); p = 0.001] were more evident as predictors of mortality (Table 6).
5. Discussion
The current study is one of the few to have explored the course of androgen deprivation in 51 patients treated for PCa Apart from the response to this hormonal deprivation, the objective was to evaluate the survival of the patients and to look for the different predictors of mortality. In Table 1, it was from the fourth decade that PCa was diagnosed in our patients with an average age of 69.4 years (40 - 92 years). Many studies report either an average around the sixth decade [3] [14] [15] [16] [17] or around the seventh decade [18] - [24] with extremes that do not show significant differences. Only one reports an average of 59.13 years [25] (Table 7).
Hypertension was the most common comorbidity in 25.5% of cases. Studies have explored the association of PCa with hypertension as comorbidity and report different results. Some report hypertension as the only comorbidity; 33.3% [17] and 35.4% [26] cases. Others show an increased risk of PC [27], death [28], and an increased prevalence of PCa in hypertensive Africans [29]. Most of the patients were carriers of PCa diagnosed at the stage of metastasis followed by cases at high risk of progression and intermediate risk. Those of low risk have been rare. The same results are described in the literature, with the predominance of metastatic cancers for some authors [15] [21] [24] [29] [30]. Others report more on low-risk and intermediate-risk cancers [31] [32] [33]. Indifferent types of studies. Bone metastases predominated, followed by multiple locations and lymph nodes. PCa is first recognized as osteophytes cancer before any other
Table 6. Distribution of patients according to the predictors of mortality.
Table 7. Char of mean age according to authors.
NR: No reported.
localization [21] [34] [35] [36] [37] [38]. We used clinical signs, PSA and testosterone level, cTNM stage, Gleason score, and medical imaging to assess castration resistance in our patients. Everything was summed up in D’Amico’s classification. Numerous studies have explored similar parameters in various ways to assess castration resistance in PCa [39] [40] [41]. Most patients had benefited from castration (hormonal or surgical), radical prostatectomy was very rare (3.9%) (Table 2). These results are almost like those found in the literature [20] [42] [43] [44]. The castration-resistant cancer rate (CRPC) was 43.1% within a median of 1.4 (1 - 3 years) year (17 months) of response to treatment (Figure 1). Our results are different from those of other authors [11] [34] [45] [46] [47] [48]. Other studies [49] report that 10% to 20% of PCa evolve into CRPC approximately 5 years after the start of treatment (Table 8).
Gleason score 8 - 10 (log-rank; p = 0.018) (Figure 1), High-risk Cap [OR 2.95 95% CI (1.36 - 4.69)] (Table 3), metastatic [OR 5.88 95% CI (1.62 - 7.99)] (Table 3), and dysuria (Table 4) were providers of CPRC. These same results are reported by many researchers [11] [21] [34] [44] [49] [50] [51] [52]. The increase in PSA level was influenced by clinical stage cT3 - cT4 [OR 15.0 95% CI (2.02 - 17.11); p = 0.001], and Gleason score 8 - 10 [OR 6.07 95% CI (1.49 - 24.76); p = 0.035] (Table 5). These same results are repeated in many studies [53] - [58]. In terms of percentage, patient survival was 94.7%, 88.0%, 89.7%, and 80.4% at 1 year, 2 years, 3 years, and 4 years, respectively. Three groups of auteurs report report different survival percentages. The first indicates a survival at 1 year, 2 years, 3 years, and 4 years between 80% to 90% [16] [21], the second suggests a 5-year survival of 30% for CRPCs [55]. In the end, the Henry Botto team evokes a survival of 21.1%% at 8 years [20].
In terms of months or years, the median patient survival was 30 [24] - [30] months and the mean survival was 26.6 months. Most studies already published report a median survival that varies between 14 to months, with certain differences depending on the stage, grade, and comorbidity [16] [18] [19] [20] [24] [34] [53] [59] [60]. Patients resistant to castration (LogRank, p = 0.029) (Figure 2), those with a PSA level ≥ 100 ng/ml (LogRank, p 0.006) (Figure 3) or a Gleason score 8 - 10 (LogRank, p 0.004) (Figure 4) had significantly reduced survival compared to the others. Other researchers come to the same conclusion [11] [21] [38] [61]. Searching for predictors of mortality; in univariate analysis; PSA levels ≥ 100 ng/ml [HR 10.20 95% CI (1.29 - 13.56); p = 0.001], Gleason score 8 - 10 [HR 10.97 98% CI (1.39 - 16.68); 0.035] resistance to castration [HR 3.98 95% CI (1.56 - 7.04); p = 0.017] and metastases [HR 2.67 95% CI (1.69 - 10.35); p = 0.007] were predictors of mortality, without any significant difference within each group. In multivariate analysis, the Gleason score 8 - 10 [HRa 10.15 95% CI (2.18 - 12.23); p = 0.035] and the PSA level ≥ 100 [HRa 8.49 95% CI (2.15 - 10.56) p = 0.001] (Table 6), were more prominent as predictors of mortality.
Table 8. Chart of castration resistance according to the authors (34).
The same observation is made by many authors namely; the mortality is all the higher for high-grade, metastatic PCa, RCPC and a very high PSA rate [21] [36] [51] [61] [62] [63].
6. Conclusion
Prostate cancer is a public health challenge in our area. The average age of the patients was 69.4 years (40 - 92 years). Dysuria was the main symptom of medical consultation. Hypertension was the main comorbidity among our patients. Most of our patients have been diagnosed at metastasis stage or a high-risk stage of progression. Castration resistance was observed from the 5th month of treatment, especially for patients with metastases. The median patient survival was 30 months and the average survival was 26.6 months with a difference depending on the stage, grade, and comorbidity. The overall mortality rate was 19.6%. The patients with castration resistance had significantly reduced survival.
Limitations of the Study
This publication is considered as a pilot study which will be validated by others. The interpretation of the results should consider the limitations. The lack of randomization of the subjects studied and the limited to one center introduced a selection bias and does not allow the generalization of our results. The retrospective nature of our cohort constitutes a significant loss of some useful information. The low socio-economic level of the patients made it impossible to carry out several paraclinical examinations which have a definite influence on the survival of patients. Finally, the relatively small size of the sample could not give enough power to statistical tests to detect possible associations between the variables of interest. To get around these methodological pitfalls, a multicenter randomized longitudinal cohort study is expected.
Authors’ Contribution
Dieudonné Molamba Moningo: Research design and supervision.
Junior Konga Liloku: Data collection and writing.
Richard Koseka Demongawi: French English translator.
Nkodila Aliocha: Statistical analysis.
Other authors: Corrections, remarks, and suggestions.
Appendix
DATA COLLECTION SHEETS
Subject: “EVOLUTION OF ANDROGENIC DEPRIVATION IN TREATMENT OF PROSTATE CANCER IN KINSHASA”
Date of collection at the Pointe à Pitre/Matete Clinic ………………………
March 2014 to June 2018.
I. IDENTITY
Coded …………………………………
Place and date of birth …………………………………
Weight (kg) …………….... kg
Height (cm) ……................ cm
Marital status Married: ☐
Divorced: ☐
Single: ☐
Widower:
Age (year) ……………. years
Address C: ……………………………...
Q: ………………………….......
Av: ………………………….....
Profession ......................................................
Province of origin .....................................................
Phone number .....................................................
II. MEDICAL HISTORY
1) Personal
- Smoker Yes ☐ No ☐
- Former smoker Yes ☐ No ☐
- How many stems/day ......................................................
- Allergy to a drug Yes ☐ No ☐
- (which)? ......................................................
- Hypertensive Yes ☐ No ☐
- Diabetic Yes No
- Fracture, in the absence of major trauma
(If yes, which fracture site) ........................................................
- Hyperuricemia Yes ☐ No ☐
- Alcohol Yes ☐ No ☐
2) Family history
- Prostate cancer Yes ☐ No ☐
- Breast cancer Yes ☐ No ☐
III. CLINICAL FINDING
Dominant symptom .........................................................
Functional signs: 1. pain:
- Pain in RT Yes ☐ No ☐
- Bone pain Yes ☐ No ☐
2) Urinary disorder:
- dysuria Yes ☐ No ☐
- incontinence Yes ☐ No ☐
- mictalgia Yes ☐ No ☐
- pollakiuria Yes ☐ No ☐
- nocturia Yes ☐ No ☐
- resumes post-voiding Yes ☐ No ☐
3) Rectal touch:
- nodule Yes ☐ No ☐
- invasion Yes ☐ No ☐
- lumbar contact Yes ☐ No ☐
- Lower limb edema Yes ☐ No ☐
IV. PARACLINIC
Imaging:
- endorectal ultrasound Yes ☐ No ☐
- MRI Yes ☐ No ☐
- Abdomino-pelvic scanner Yes ☐ No ☐
- bone scan Yes ☐ No ☐
Prostate biopsy:
- Performed Yes ☐ No ☐
- Positive result Yes ☐ No ☐
Method of realization:
- echoguided Yes ☐ No ☐
- transrectal Yes ☐ No ☐
Initial stage
- localized cancer Yes ☐ No ☐
- locoregional cancer Yes ☐ No ☐
- metastatic cancer Yes ☐ No ☐
Clinical classification/Imaging, cTNM cT......N......M.......
Gleason score .....................................................
Testosteroneemia (ng/ml) (1.......), (2........), (3........), (4.......)
PSA (ng/ml) (1.......), (2........), (3........), (4.......)
D’AMICO classe: Low risk (1), Intermediate risk, (2) High risk (3), Metastatic (4)
V. TREATMENT
1) Hormone therapy used R/......................................................
R/......................................................
R/.......................................................
Start of treatment (month/year) .........................................................
Duration .........................................................
2) Anti-androgen therapy? Yes ☐ No ☐
Which? R/......................................................
R/......................................................
R/......................................................
3) Surgical castrations .........................................................
.........................................................
Others R/.....................................................
Complications:
- Early .........................................................
.........................................................
-late .........................................................
.........................................................
Evolution of castration .........................................................
.........................................................
.........................................................
Evolutions of testosterone .........................................................
.........................................................
.........................................................
Time to onset of castration resistance after treatment is indicated
..........................................................
..........................................................
Death Yes ☐ No ☐
Death with cancer Yes ☐ No ☐
Age: .................................................
Time after treatment......................
Other causes ..........................................................