Efficacy of switching therapy of luteinizing hormone–releasing hormone analogue for advanced prostate cancer

Abstract

This study was conducted to determine the efficacy of switching therapy with a second‐line luteinizing hormone–releasing hormone (LHRH) analogue after prostate‐specific antigen (PSA) progression for advanced prostate cancer. We enrolled 200 patients, from December 2005 to September 2013, with nodal positive, metastatic prostate cancer or disease progression after definite treatment receiving continuous LHRH analogue therapy with monthly depot leuprorelin(sc) acetate 3.75 mg/vial (LA) or goserelin acetate(sc) 3.6 mg/vial (GA). If the patients had castration‐resistant prostate cancer, the treatment choice of switching therapy (from LA to GA or from GA to LA) prior to starting chemotherapy was given. The LH, testosterone level, and PSA change were recorded. The records showed that there were 127 patients receiving LA as initial ADT therapy, whereas the other 73 patients were in GA therapy. A total of 92 patients received LHRH analogue switching therapy (54 patients switched from LA to GA and 38 switched from GA to LA). The effect of LH and testosterone reduction prior to and after switching therapy was comparable between the two groups, and increased PSA level after 3 months of treatment was seen in both groups (median PSA: 15.7–67.7 ng/mL in the LA to GA group; 15.2–71.4 ng/mL in the GA to LA group). This study concluded that switching therapy for patients with PSA progression after ADT has no efficacy of further PSA response.

Introduction

Androgen deprivation therapy (ADT) with luteinizing hormone–releasing hormone (LHRH) analogue has been the standard treatment for advanced prostate cancer since the demonstration of the hormone dependence of prostate cancer cells. In the past 20 years, medical castration with the use of LHRH analogue has been proved as a tolerated and effective option for advanced prostate cancer because of its survival benefits [[1], [2], [3], [4]]. It has also become the replacement for surgical castration (bilateral orchiectomy) because of its convenience and avoidance of body integrity loss.

ADT with LHRH analogues decreases PSA and improves clinical symptoms in about 90% of patients. However, limitation in the duration of response for the drug exists and after that castration‐resistant prostate cancer (CRPC) develops. Different LHRH analogues are considered equivalent [[4], [5], [6]], so there was no rule about changing one LHRH analogue to another in the former concept prior to starting therapy for CRPC. However, various agents have different chemical structures, so it is possible that they could result in different patient responses. A study from multiple institutions in Canada demonstrated the efficacy of second‐line LHRH analogues after disease progression [7]. They showed PSA decrease after switching therapies, and the decrease was more significant in the group switching from leuprolide acetate (LA) to goserelin acetate (GA). According to the hypothesis of divergent way of action from various LHRH analogues and the possible efficacy of switching therapy, we performed this study to determine if changing LHRH analogues would lead to PSA response.

Methods

Patient selection, treatment, and follow‐up

We performed a prospective study from December 2005 to September 2013 in a single medical institute. We identified patients with initially diagnosed nodal positive (N1), metastatic (M1) prostate cancer and disease progression after definite treatment including radical prostatectomy and radiotherapy. The patients received continuous LHRH analogue therapy as the initial ADT with Depot leuprorelin(sc) acetate 3.75 mg/vial (LA) or goserelin acetate(sc) 3.6 mg/vial (GA) every month, and the concomitant nonsteroidal antiandrogen was only allowed for 1–4 weeks when we initiated ADT to avoid flare‐up phenomenon. Other patients receiving hormonal manipulation such as diethylstilestrol, steroidal antiandrogen, or ketaconazole were excluded. Furthermore, if patients had met the criteria of CRPC, defined as two consecutive PSA increases while on ADT and reaching castration testosterone level (< 0.2 ng/mL), we gave these patients the choice of switching therapy from the original LHRH analogues to another regimen (LA to GA or GA to LA) prior to starting chemotherapy treatment. The second‐line LHRH analogue was used for at least 3 months to determine the PSA response. Baseline data including patient demographics, initial TNM stage, baseline PSA, and Gleason score were collected. The LH, testosterone, and PSA level were checked every 3 months and whether patients could reach the cutoff value (LH < 1.0 mIU/mL and testosterone level < 0.2 ng/mL or <0.1 ng/mL) was determined at the first 3‐month follow‐up. The PSA level prior to switching therapy and every 3 months after the therapy were recorded to analyze the treatment response. This study was approved by the CGMH Institutional Review Board.

Statistical analysis

The baseline demographic data of patients and the efficacy of lowering hormone level were analyzed with Fisher's exact test, Pearson's Chi‐square test, and two‐tailed independent t test. McNemar test, Wilcoxon signed rank test, and Mann–Whitney U test were used to show the significance of LH or testosterone change and PSA response. All analyses were carried out using SPSS software, version 17.0 (SPSS Inc., Chicago, IL, USA).

Results

From December 2005 to September 2013, a total of 200 patients met the inclusion criteria. The baseline data of patients are shown in Table 1. Overall, 127 patients received LA as the initial ADT therapy and the other 73 patients were in GA therapy. There was no statistical difference in terms of age, initial TNM stage, Gleason score, or baseline PSA between the groups of patients receiving LA or GA. The efficacy of reducing LH less than 1.0 mIU/mL (73.2% in LA and 74% in GA group; p = 0.909) and testosterone less than 0.2 ng/mL (81.9% in LA and 80.8% in GA group; p = 0.851) was equivalent. As regards the extremely low testosterone level of less than 0.1 ng/mL (the lowest sensitive level in our clinical laboratory according to the new rapid polyelectrolyte‐based immunofiltration technique [8]), there were 49.6% patients in the LA group and 47.9% patients in the GA group who could reach this level.

Table 1.

Baseline patient characteristics.

		Leuprorelin No. (％)	Goserelin No. (％)	p
N		127	73
Age (y)		77.1 ± 8.5	76.7 ± 8.4	0.757
Initial PSA (ng/mL)		404.1 ± 1293.2	583.3 ± 1689.2	0.413
Gleason score	2–4	2 (1.8)	1 (1.6)	0.249
	5–7	54 (49.5)	24 (38.1)
	8–10	53 (48.6)	38 (60.3)
Initial T stage	1	9 (9.1)	4 (7.4)	0.198
	2	22 (22.2)	5 (9.3)
	3	52 (52.5)	36 (66.7)
	4	16 (16.2)	9 (16.7)
Initial stage	1	5 (5.2)	1 (1.9)	0.113
	2	16 (16.7)	3 (5.8)
	3	34 (35.4)	17 (32.7)
	4	41 (42.7)	31 (59.6)

PSA = prostate‐specific antigen.

Switching therapy

In the study population, there were 92 patients included in the switching therapy with the available clinical data. Fifty‐four patients switched from LA to GA and 38 patients switched from GA to LA because of PSA progression. In the univariate analysis, there was no statistically significant difference in patient age, initial Gleason score, and duration of first LHRH analogue usage. There was also no statistically significant difference in median PSA prior to the first or secondary LHRH analogue between the two groups (Table 2). After rechallenging with the second‐line LHRH analogues, the effect of LH and testosterone reduction was the same between the two groups (Table 3). About PSA, the increase in level after 3 months of treatment was seen in both groups (median PSA: 15.7–67.7 ng/mL in the LA to GA group and 15.2–71.4 ng/mL in the GA to LA group; Table 4) and the result of absolute PSA change showed no significant difference (p = 0.341). There were 36 patients (17 patients in the LA to GA group and 19 patients in the GA to LA group) who kept second‐line LHRH analogue without definite CRPC treatment and receiving the 6‐month PSA follow‐up. The PSA level continued to rise (204.2 ng/mL and 174.4 ng/mL) at the 6‐month follow‐up.

Table 2.

Univariate analysis of switching therapy.

	Overall	LA to GA	GA to LA	p
Patient No. (%)	92	54	38
Mean age at PC diagnosis (y)	77.9 ± 7.5	77.6 ± 7.6	78.4 ± 7.5	0.591
Gleason score (%)
≤ 6	18	8	10	0.171
≥ 7	74	46	28
Mean duration of first analogue (mo)	30.0 ± 19.5	33.9 ± 22.2	24.3 ± 13.1	0.11
Median PSA (ng/mL)
At first LHRH analogue start (range)	14.4 (3.42–947.8)	28.2 (3.42–947.8)	8.76 (4.1–698.5)	0.087
At second LHRH analogue start (range)	15.8 (3.05–698.46)	15.7 (3.05–213.29)	15.2 (4.05–698.46)	0.701

GA = goserelin acetate; LA = leuprorelin acetate; LHRH = luteinizing hormone–releasing hormone; PC = prostate cancer; PSA = prostate‐specific antigen.

Table 3.

Hormonal suppression effect of switching therapy.

Primary medicine		Secondary medicine		p a
		Goserelin	Goserelin
	LH	<1.0 mIU/mL	≥1.0 mIU/mL
Leuprorelin	<1.0 mIU/mL	28	11	0.481
Leuprorelin	≥1.0 mIU/mL	7	8
		Leuprorelin	Leuprorelin
	LH	<1.0 mIU/mL	≥1.0 mIU/mL
Goserelin	<1.0 mIU/mL	26	8	0.109
Goserelin	≥1.0 mIU/mL	2	2

Primary medicine		Secondary medicine		p a
		Goserelin	Goserelin
	Testosterone	<2.0 ng/mL	≥2.0 ng/mL
Leuprorelin	<0.2 ng/mL	48	1	>0.99
Leuprorelin	≥0.2 ng/mL	0	5
		Leuprorelin	Leuprorelin
	Testosterone	<2.0 ng/mL	≥2.0 ng/mL
Goserelin	<0.2 ng/mL	29	0	>0.99
Goserelin	≥0.2 ng/mL	0	9

LH = luteinizing hormone.

^aMcNemar test.

Table 4.

PSA response of switching therapy.

	N	Median PSA (IQR, ng/mL)				Absolute PSA change (3 mo & 0 mo)	p a
		At second LHRH analogue start (0 mo)	3 mo after second LHRH analogue	N	6 mo after second LHRH analogue
Leuprorelin to goserelin	54	15.7 (5.4, 69.1)	67.7 (10.8, 156.4)	17	204.2 (21.5, 320.4)	34.2	0.085
Goserelin to leuprorelin	38	15.2 (4.8, 149.2)	71.4 (8.0, 134.5)	19	174.4 (24.3, 207.3)	28.2	0.918
						p = 0.341 b

IQR = interquartile range; LHRH = luteinizing hormone–releasing hormone; PSA = prostate‐specific antigen.

^aWilcoxon signed rank test.

^bMann–Whitney U test.

Discussion

Based on the dependence of prostate cancer cells on testosterone, surgical or chemical castration is used for stopping the growth of cancer cells. When choosing the method of castration, the goal is to achieve a low testosterone level. Historically, testosterone level below 0.5 ng/mL was used as the standard level of castration [9]. However, after the development of new assays and more sensitive tests, the level to achieve castration seems to be more strictly defined [[8], [10]]. Recent studies have measured the testosterone level after bilateral orchiectomy based on chemiluminescence and reported the mean value to be about 0.2 ng/mL [[3], [11], [12], [13]]. Therefore, < 0.2 ng/mL was assumed to be an adequate castration level when using chemical castration [14]. Oefelein and Resnick [15] advocated for the concept of “the lower, the better” while defining the level of castration—which is why we analyzed the efficacy of lowering testosterone not only by 0.2 ng/mL but also by 0.1 ng/mL (the lowest sensitive level in our institute).

In this study, about 80% of patients with nodal positive, metastatic disease or disease progression after definite therapy receiving LA or GA could achieve a satisfactory level of testosterone. About 70% patients could have an LH level of less than 1 mIU/mL. Because the pattern of change in LH was similar to that in testosterone, studies seldom showed the level of LH [[16], [17]]. Whichever regimen is used for ADT, the efficacy of hormonal control was similar and the percentage of patients who could attain the lowest testosterone level (< 0.1 ng/mL) was around 50% in both groups. We found that a proportion of PC patients (∼20%) did not reach the goal of castration with testosterone of less than 0.2 ng/mL, and the other series also revealed similar results, in which about 10–37% failed to achieve low testosterone level [[18], [19], [20], [21]]. In general, castration level can be achieved 4 weeks after the treatment [13], and the duration of testosterone suppression ranges from 9.3 months to 45 months [[1], [3], [13], [16]]. Different series showed variable duration because of variable baseline conditions. Nadir PSA, time to nadir PSA, baseline PSA, tumor grade, and pretreatment testosterone are predictive of patient response.

For these patients with PSA progression, switching to another LHRH analogue or adding other secondary hormonal agent was offered as a treatment option. In our setting, some CRPC patients hesitated to proceed to chemotherapy because of their fear of side effects and the inconvenience of hospitalization, so we offered them another alternative treatment with second‐line LHRH analogues prior to chemotherapy. Lawrentschuk et al [7] reported a second‐line LHRH analogue having the efficacy of PSA decrease in 70% patients, and the duration of response after switching was about 5 months. They also found that the trend was more significant in the group that switched from LA to GA. They advocated the role of second‐line LHRH analogues. However, only retrospective evaluation was performed, and no testosterone or LH data were made available. Furthermore, no other series has so far duplicated their result.

Increased PSA after ADT generally indicates disease deterioration. Some experts doubt if the failure of the initial LHRH analogue is attributable to the mutant castration‐resistant cell development. They assume that PSA progression ensues only because of the tolerance to one LHRH analogue, so another similar drug with different chemical structures may result in PSA decrease again. Many questions exist regarding the optimal dosing, duration, and route of administration of LHRH analogues [[22], [23], [24], [25], [26]]. The dose of leuprorelin used in this study is only 3.75 mg, in comparison with 7.5 mg used in the United States. We still cannot find any difference in androgen suppression between leuprorelin and goserelin. Other series also showed similar results [[27], [28], [29]].

In our study, we did not find any indication that second‐line LHRH analogues resulted in further decrease in PSA even under equivalent low LH and testosterone levels prior to and after switching therapy. Although different mechanisms of action of different drugs may exist, there was no proven effect on PSA response and survival by rechallenging with alternative LHRH analogues. The limitation of our study was the small case number and outcome such as progression‐free survival or overall survival was not shown because of the variable of subsequent therapy. A potential benefit—whether in survival or symptom relief—of second‐line LHRH analogues may exist in different subsets of patients. Further prospective randomized study is necessary to determine its efficacy.

In conclusion, this study revealed the equivalent effect of LA and GA in initial testosterone control for advanced prostate cancer. In subsequent switching therapy for patients with PSA progression after ADT, there was no further PSA response after switching to secondary LHRH analogue.