ERG Rearrangement For Predicting Subsequent Cancer Diagnosis Of High Grade Prostatic Intraepithelial Neoplasia And Lymph Node Metastasis And The Surgical Treatment For Locally Advanced Prostate Cancer

Background and Objective:Prostate cancer is the most common urological malignancy in Western countries, which wasthe second leading cause of death following lung cancer. The incidence of prostate cancer isrelatively low in China. But with the changes of diet structure and living habits since the lastdecade, the incidence of prostate cancer is increasing in our country. In2009, the incidence ofprostate cancer in Shanghai was about25.88/100,000. It indicates that prostate cancer hasgradually become one of the major diseases that threaten the health of our male population.The final diagnosis of prostate cancer must rely on the prostate biopsy. Our previous studyanalyzed the clinical and pathological data of923cases with serum PSA level of4-20ng/ml whounderwent12needle prostate biopsy schemes in the past10years. The results showed that overallcancer detection rate was27.4%. There were18.3%,3.5%and50.8%were diagnosed ashigh-grade prostatic intraepithelial neoplasia (HGPIN), Atypical small acinar proliferation ofprostate (ASAP) and benigen prostatic hyperplasia (BPH), respectively. Small amount of prostate tissue is collected during needle biopsy. And the conventionalimaging methods can not accurately locate the tumor lesions. So we can not precisely analyze anddiagnose a small amount of tumor cells mixed with a large number of normal cells or glands.Especially, it is hardly to distinguish early focal prostate cancer and the HGPIN. There are about20%patients who received prostate needle biopsy for elevated serum PSA level dignozed asHGPIN. HGPIN is considered to be precancerous lesions of the prostate cancer. However, onlyabout20-40%of those diagnosed as PIN at initial biopsy are diagnosed as prostate cancer insubsequent repeat biopsy. These data indicates that not all patients with HGPIN are required toaccept re-biopsy urgently. Several researches demonstrate that biochemical and molecularabnormalities appear prior to the changes of morphology features for many malignant tumors.They suggest that molecular pathology methods can be used as a supplementary measure toidentify these early cancers.Tomlins and colleagues found that TMPRSS2-ERG rearrangementwas a prostate cancer-specific molecular marker. We intend to detect ERG rearrangement ratesusing fluorescence in situ hybridization (FISH) in biopsy samples which were diagnosed asHGPIN at initial biopsy. Patients were divided into two groups according to the cut-off (1.6%)established in our previous research using receiver operating characteristic curve (ROC). All theenrolled HGPIN patients received one repeat biopsy within3-6months after initial biopsy. Andthen, re-biopsy was done on the base of PSA velocity. Cancer detecting rates were compared inthese two groups of HGPIN patients.Lymph node metastasis is a leading prognositic factor for prostate cancer. However, it is verydifficult to accurately assess lymph node metastasis of prostate cancer before surgery. Magneticresonance image (MRI) and computed tomography (CT) were the most common used modality todetect lymph node metastases in recent clinical practice, but its sensitivity is less than40%. Somenomograms involved serum PSA level, biopsy Gleason score and clinical T stage have beenintroduced to predict lymph node metastasis of prostate cancer in recent years. The sensitivity andspecificity of these models are also only about70.0%. Thus, some robust methods with highsensitivity and specificity are urgently needed to deal with this clinical challenge. Recent studiesdemonstrate that only those with positive ERG rearrangement prostate cancer locus had potentialability to form a metastatic lesion. It indicated that TMPRSS2-ERG might be employed to predictlymph node metastasis of prostate cancer. In current study, we plan to evaluate ERG rearrangement rates in patients underwent prostatectomy and pelvic lymph node dissection.Optimal cut-off for predicting lymph node metastasis of prostate cancer is established using ROCcurve. We aim to introduce a new method for predicting lymph node metastasis of prostate cancer.Clinical localized prostate cancer might be upstaged as T3-T4advanced prostate cancer aftersurgery. According to the previous concept, these patients are not suitable for radical surgery.Radiotherapy combined with hormonal therapy or conservative endocrine therapy alone is thestandard treatment strategy for this subgroup of patients. Recently, a number of studies haveshown the efficacy of radical surgery for the treatment of locally advanced prostate cancer. And itsefficacy is as well as radiotherapy, and even better than endocrine therapy alone. In addition,remove of the primary prostate cancer focus can increase the response of adjuvant endocrinetherapy, systemic metastatic can be delayed. Consequently, we begin to re-discuss the radicalsurgery for locally advanced prostate cancer. This study retrospectively analyzes the clinical andpathological characteristics of152patients with pathological advanced prostate cancer whounderwent laparoscopic radical prostatectomy in our center. Function recovery and long-termtumor control results are recorded and analyzed. We aim to provide further clinical practice of thesurgical treatment of this group of patients.Materials and Methods:1、On the basis of the cut-off value established previously, the162patients with HGPIN werestratified to two groups: one group with positive ERG rearrangement and the other with negativeERG rearrangement. Subsequent prostate cancer detecting rates between the two groups werecompared.2、One hundred and forty-three biopsy samples from whom underwent prostatectomy andpelvic lymph node dissection were collected. ERG-rearrangement rate was evaluated using FISH.The optimal cut-off of ERG rearrangement for predicting lymph node metastasis was establishedusing receiver operating characteristic curve (ROC). And the sensitivity and specificity of thiscut-off were calculated.3、From January2004to July2011,152cases including105locally advanced PCa and47lymph node metastatic PCa who treated by LRP with extended lymph node dissection (ePLND)were included in this study. Operation records, complications and urinary continence rates werepresented. Oncologic outcomes expressed as3-and5-yr biochemical recurrence-free, overall andcancer-specific survival rates were analyzed using Kaplan-Meier survival curve.Results:1. Of the162HGPIN patients,59HGPIN with positive ERG rearrangement and103with negative ERG rearrangement were observed. A total of56of59(94.9%) HGPIN cases withpositive ERG rearrangement and5of103(4.9%) HGPIN cases with negative ERG rearrangementwere diagnosed with prostate cancer during repeat biopsy or transurotheral resection of prostate(TURP). The rates of subsequent detection of prostate caner between the two groups werestatistical different (p<0.001).2. Of the143patients who underwent prostatectomy and pelvic lymph node dissection in ourinstitute from January2004to July2011,56patients had positive lymph node and87cases hadnegative lymph node. There were significant differences in rearrangement rates of ERGrearrangement between lymph node-positive and-negative prostate cancer (P <0.001). Theoptimal cutoff value to predict lymph node metastasis by ERG rearrangement was established,being2.6%with a sensitivity at80.4%[95%CI:67.6–89.8] and a specificity at85.1%(95%CI:75.8–91.8). The positive predicting value and negative predicting value were0.776（95%CI：0.668-0.853） and0.871（95%CI：0.791-0.923）, respectively。3. The mean operation time and bleeding were240minutes and110ml, respectively. After12-87months (median48months) of follow-up,91.4%and94.7%of the patients were urinarycontinence at6and12months, respectively.80biochemical recurrent diseases were detected. The3-and5-yr biochemical progression-free survival rates were59.2%and47.3%, respectively.Multivariate analysis showed that Gleason score (HR:1.66,95%CI:1.05-2.64, P=0.031),pathological stage (HR:1.64,95%CI:1.2-2.23, P=0.002) and surgical margin (HR:1.75,95%CI:1.04-2.95, P=0.035) were independent risk factors for subsequent biochemical relapse. The3-and5-yr overall and cancer-specific survival rates were90.2%,86.0%and95.8%,92.3%, respectively.There were no significant differences in biochemical recurrence-free (42.6%vs49.5%, P=0.491),overall (83.4%vs87.3%P=0.503) and cancer-specific survival rates (92.3%vs94.9%, P=0.801)between lymph node positive and negative PCa.Conclusion:1、HGPIN patients with positive ERG rearrangement had significant higher rate for subsequentdiagnosis of prostate cancer in re-biopsy than those with negative ERG rearrangement. Forpatients with positive ERG rearrangement, prostate cancer related therapy can be adapted forrather high incidence of prostate cancer detection rate, while those with negative ERGrearrangement HGPIN, the probability of detecting prostate cancer is low enough to omit rebiopsy.2、ERG rearrangement with the optimal cut-off of2.6%with both high sensitivity and specificitycan accurately predict lymph node metastasis of prostate cancer. PLND can be conductedaccording to this promising biomarker.3、LRP plus ePLND is feasible to cases with locally advanced non-extra node metastatic PCa.