Prostate-specific antigen (PSA) is considered to be an effective prostate tumor marker. The value of PSA as a prostate tumor marker is exhibited particularly well when used to monitor patients following treatment, wherein the majority of follow-up analysis is determined by PSA alone. Specifically, the PSA nadir value of patients that underwent a radical prostatectomy (RP) becomes theoretically zero or demonstrates a very low concentration due to the lack of prostate tissue.

By contrast, the PSA nadir value, which indicates the limit of sensitivity of the measurement kit used, theoretically demonstrates the presence of residual prostate tissue in patients exhibiting benign or malignant disease, including distant metastasis.

The detection limit of recent PSA kits has increased when compared with conventional PSA kits; a third-generation high-sensitivity PSA kit, which was successfully used in a study by Yu et al (1) was adopted at the National Kyushu Cancer Center (Fukuoka, Japan) thereby enabling PSA detection at levels <0.008 ng/ml. This enabled determination of the presence of particularly small prostate tissue remains in patients following RP, which usually cannot be detected by computed tomography, magentic resonance imaging or bone scintigraphy from PSA values.

Numerous clinicopathological factors have been proposed to predict biochemical failure (BCF) following RP, including a positive resection margin (RM1), preoperative PSA, clinical stage, Gleason score, extraprostatic extension, seminal vesicle invasion, positive lymph nodes and positive surgical margins among others (2–4). Many of these clinicopathological factors may be associated with biological aggressiveness and potentially with distant metastasis.

Theoretically, a RM1 indicates that there are prostate cancer remains, which affect the PSA value. However, even when the resection margin is positive, there are numerous cases in which the postoperative PSA value declines to less than the measuring limit value and BCF is not detectable. Therefore, to demonstrate that the resection stump of an excised specimen does not necessarily indicate the true resection margin, a retrospective investigation was conducted using the clinicopathological data of patients that underwent an RP, but had not received preoperative treatment.

No other tumor marker has been determined to be as efficacious in the diagnosis, management and treatment of a disease as the PSA level in the setting of prostate cancer.

The ectopic expression of PSA has been reported in smaller concentrations in the tissue of malignant breast tumors (10), normal breast tissue (11,12), breast milk (13), female serum (12), and adrenal and renal carcinomas (14); however, for practical and clinical purposes, PSA is organ-specific and is primarily produced by prostatic luminal epithelial cells (13,15,16). Although it is organ-specific, PSA is not cancer-specific, as demonstrated by the substantial overlap in values observed between males with benign versus malignant prostate disease (17,18).

The PSA level is expected to be undetectable within six weeks of a successful RP (20). Thus, a persistently elevated PSA level indicates the presence of PSA-producing tissue within the body. In patients treated with RP, such findings are generally hypothesized to reflect residual cancer due to the presence of either micrometastases (that were not detected or were undetectable prior to surgery) or residual disease in the pelvis, potentially due to an RM1 (9). Yu et al (1) classed reagents with an ultrasensitive PSA detection threshold value of 0.1–0.3 ng/ml as first-generation agents, those with a value of 0.02–0.1 ng/ml as second-generation agents and those with a value of 0.001–0.02 ng/ml as third-generation agents. Due to the development of such a highly sensitive PSA reagent, even smaller values became measurable following RP when compared with conventional reagents, thereby enabling clinicians to observe the presence of PSA-producing tissues regardless of whether they are benign or malignant and even when PSA is present in particularly small quantities.

At the National Kyushu Cancer Center, ultrasensitive PSA was used during the follow-up period in patients that had undergone an RP between September 2003 and March 2011, with a measuring limit value of <0.008 ng/ml. Accordingly, when the value drops to less than the postoperative measuring limit, it may be proposed that the PSA-producing tissues do not theoretically exist, or only particularly small PSA-producing tissues (<0.008 ng/ml) are present.

As shown in Table I, the PSA nadir fell to <0.008 ng/ml in 239 cases (86.9%) among the 275 patients undergoing RP in the present study, while BCF was noted in 43 cases (15.6%). In a previous study, Ellis et al (21) reported the PSA value to be below the measurement sensitivity level of 0.008 ng/ml in 86.2% of cystoprostatectomy cases in which prostate cancer was not detected, thus indicating that RP was being performed appropriately at the National Kyushu Cancer Center.

Generally, in surgical oncology, a wider excision margin is associated with a lower likelihood of residual disease. This concept may be applied to organs that are present in abundance, such as the bowel. In urology, for example, Khalifeh et al (22) reported that the cases exhibiting positive surgical margins, among patients treated via robot-assisted partial nephrectomy, displayed higher rates of recurrence and metastasis (P<0.001).

As indicated in Table IIA, univariate analysis demonstrated RM1 as a significant factor affecting BCF (P=0.0005). However, according to the multivariate analysis (Table IIB), significant factors affecting BCF included the preoperative PSA levels, pT, RP Gleason score, and PSA nadir, with RM1s not identified as a significant factor affecting BCF. A total of 56 of 275 cases (20.4%) were RM1, among which BCF was observed in only 17 cases (30.4%; Table I).

According to a study by Epstein et al (23), although surgical margin is a significant factor, the recurrence rate is ~30–50% even when the margin is positive; however, not all cases recur, with no significant difference from the results of the present study. Therefore, a comparative investigation was conducted for the PSA nadir value in each RM1 case. As demonstrated in Table III, the postoperative PSA value decreased to less than the measurement sensitivity (0.008 ng/ml) in 41 cases (73.2%) out of all of the RM1 cases. A comparative investigation was initially performed between the PSA nadir <0.008 ng/ml and the PSA nadir ≥0.008 ng/ml groups from among all 42 cases (75.0%) in which only one site was RM1; however, there was no significant difference observed between RM1 location in the two groups (apex, P=0.1460; base, P=0.1384; anterior, P=0.3870; and posterolateral, P=0.5040). Subsequently, a comparative investigation was conducted between the PSA nadir <0.008 ng/ml and the PSA nadir ≥0.008 ng/ml groups for the quantity of sites that were RM1 (one vs. multiple); however, no significant difference was observed (P=0.6090). Therefore, it is proposed that an RM1 exhibited by an RP specimen may not truly reflect an RM1 within a patient.

During surgery, care must be taken when extracting a tumor, such that it does not become RM1. For example, in the area of urology, with regard to radical nephrectomies for renal cancer, resections are conducted with sufficient attention given to Gerota’s fascia, which is the loose connective tissue that connects to the renal cancer. In addition, in the case of a partial nephrectomy, the resection is conducted concurrently with ascertaining the border of the tumor during the surgical procedure, such that it does not become margin-positive, by leaving a certain degree of surgical margin. This is due to the hypothesis that when a resection margin is histopathologically diagnosed as positive, the remaining cancer tissue in the distal side of the resection margin may lead to recurrence (24). A feature of prostate cancer is the presence of multiple adenocarcinoma foci, which has been observed in 50–76% of all RP specimens in previous studies (25,26). As a result, RP is considered to be the gold standard for surgical treatment. Furthermore, surgery cannot be performed while ascertaining the border between the tumor and the normal tissues (as in renal cancer), therefore, surgery is conducted with reference to preoperative rectal examinations and imaging analysis.

In the cases described in the present study, where the tissue required removal with an enlarged margin (based on the results of a preoperative examination), the intention was to remove the prostate with adequate periprostatic tissue compared with the quantity observed in routine RP. Briefly, dissection of the dorsal vein complex was conducted adequately distally from the anterior surface of the prostate, while applying downward pressure on the prostate; in addition, wide resection of the bladder neck was performed more cranially. In the posterior region, Denonvilliers’ fascia was dissected more dorsally and in the posterolateral region, dissection of the prostatic pedicle was conducted more dorsally after separating the rectum more dorsally from the prostatic pedicle, without sparing the neurovascular bundles.

However, two elements are required to perform a successful RP against prostate cancer; the first element concerns the radicality of the cancer. As much of the surrounding tissues as possible should remain attached to the prostate to further reduce the ratio of the RM1 and thus reduce the potential for recurrence. However, the other element, which conflicts with this concept, is that function preservation (such as urinating and sexual function) is required. Therefore, the extraction range cannot be expanded and tissues may only be extracted within the range in which the urine function, in particular, may be preserved.

Theoretically, with regard to localized prostate cancer, it is hypothesized that there are two associations between the RM1 and the postoperative PSA nadir value of the excised specimen. The first is that the resection margin becomes pathologically positive due to the prostate cancer invading outside the surgical resection line, which prevents the PSA nadir value from declining to less than the measurement sensitivity. The second is that the prostate cancer does not invade outside of the surgical resection line, however, the resection margin becomes pathologically positive as a result of the surgical procedure, which leads to the PSA nadir value decreasing to less than the measurement sensitivity.

In vivo, the prostate should be surrounded by neighboring connective tissues and organs, with the prostate tissues exposed at the resection margin of the excised specimen due to its removal during the the surgical procedure. Furthermore, the useful tumor marker, PSA, may be used to understand the presence of particularly small prostate tissue remains following RP, whether they are cancerous or not.

In conclusion, regarding cases in which the postoperative levels of ultrasensitive PSA decrease to less than the measurement sensitivity (i.e. <0.008 ng/ml), although the excision specimen appears to be RM1, this may not necessarily truly indicate that the resection margin within the patient is positive.