This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Small, E. J. Articles by Eisenberger, M. Search for Related Content PubMed PubMed Citation Articles by Small, E. J. Articles by Eisenberger, M. Social Bookmarking                            What's this?

Serum Prostate-Specific Antigen Decline as a Marker of Clinical Outcome in Hormone-Refractory Prostate Cancer Patients: Association With Progression-Free Survival, Pain End Points, and Survival

From the University of California, San Francisco, CA; Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Co, Ann Arbor, MI; and Johns Hopkins University, Baltimore, MD.

Address reprint requests to Eric J. Small, MD, University of California, San Francisco, UCSF Comprehensive Cancer Center, 1600 Divisadero St, 3rd Floor, San Francisco, CA 94115; email: smalle@ medicine.ucsf.edu.

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: Validated end points are lacking for clinical trials in hormone-refractory prostate cancer (HRPC). Controversy remains regarding the utility of a posttreatment decline of prostate-specific antigen (PSA). The purpose of this study was to determine whether posttreatment declines in PSA were associated with clinical measures of improvement in a randomized phase III trial of suramin plus hydrocortisone versus placebo plus hydrocortisone.

PATIENTS AND METHODS: A total of 460 HRPC patients were randomized to receive suramin plus hydrocortisone (n = 229) or placebo plus hydrocortisone (n = 231). All patients had symptomatic, metastatic HRPC requiring opioid analgesics. Clinical end points evaluated included overall survival, objective progression-free survival (OPFS), and time to pain progression (TTPP). An evaluation of overall survival, OPFS, and TTPP as a function of a PSA decline of 50%, lasting at least 28 days, was undertaken by using a landmark analysis at 6, 9, and 12 weeks. A multivariate analysis of the impact of PSA decline was performed on these clinical end points.

RESULTS: A decline in PSA of 50% lasting 28 days was significantly associated with a prolonged median overall survival, OPFS, and TTPP, both in the entire group and the suramin plus hydrocortisone group at all three landmarks in both univariate and multivariate analysis.

CONCLUSION: In this prospective, randomized trial of suramin plus hydrocortisone versus placebo plus hydrocortisone, a posttherapy decline in PSA of 50%, lasting 28 days, was associated with prolonged median overall survival, improved median progression-free survival, and median TTPP. This analysis suggests that a posttreatment decline in PSA may be a reasonable intermediate end point in HRPC trials and calls into question the clinical utility of preclinical assays evaluating the in vitro effect of given agents on PSA secretion.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
THE LACK OF validated end points for clinical trials in hormone-refractory prostate cancer (HRPC) has long challenged clinical investigators.¹ Although improvement in survival remains the gold standard to demonstrate clinical benefit, this end point is not assessable in phase II studies and may be difficult to evaluate in phase III clinical trials, which are frequently confounded by cross-over designs used to ensure that all patients have access to potentially beneficial therapy. The lack of bidimensionally measurable disease in more than 70% of HRPC patients also precludes using objective responses in measurable disease as an end point. Posttherapy decline of prostate-specific antigen (PSA) has been evaluated in multiple reports and has been recommended as a potential marker of response.^2-4 However, the use of a decline in PSA as an intermediate marker of response has not been prospectively validated, and controversy remains as to its utility.^5,6

The use of PSA as an intermediate marker of response is further complicated by the fact that PSA levels can be affected by a variety of factors. Corticosteroids have the capacity to cause reductions in PSA and have been shown to possess palliative and antitumor properties.⁷ In addition, antiandrogen withdrawal clearly results in PSA declines in approximately 20% of patients.^8,9 Thus, some of the PSA changes noted in older reports in the literature may be attributed to antiandrogen withdrawal or the effect of concurrently administered corticosteroids.

The report that in some preclinical models, some agents result in a PSA decline that is attributed to inhibition of PSA secretion without concurrent cytotoxicity has further confounded the interpretation of posttreatment changes in PSA levels. For example, the discordance between PSA suppression and antitumor activity has been reported with suramin in preclinical in vitro and in vivo models.¹⁰ However, these laboratory observations, reported in a single publication, are of unknown clinical significance. The results of a large, placebo-controlled multicenter randomized trial comparing suramin plus hydrocortisone to placebo plus hydrocortisone have been recently reported.¹¹ The purpose of this analysis was to determine whether posttreatment declines in PSA were associated with clinical measures of improvement, including survival, time to progression, and duration of pain control for patients treated on this trial, the largest prospective phase III trial to date of systemic therapy for HRPC.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Eligibility Criteria and Treatment Plan
Eligible patients had histologically confirmed adenocarcinoma of the prostate, with painful bone metastases requiring a stable chronic regimen of opioid analgesics. The details of study design, eligibility criteria, and treatment regimen have been reported previously.¹¹ In brief, eligible patients were stratified by PSA level ( 100 ng/mL, > 100 ng/mL) and presence or absence of soft-tissue metastases and were randomized in a double-blind fashion to receive suramin plus hydrocortisone or placebo plus hydrocortisone. The therapy received by each patient was identified (unblinded) only in the event of progressive disease or dose-limiting toxicity (DLT). Progressing patients found to be receiving suramin were withdrawn from the study and observed for survival. Patients receiving placebo were eligible to enter cross-over and receive open-label suramin on the same 78-day regimen. PSA levels were measured weekly during treatment, then monthly after treatment ended. Measurable lesions were assessed at baseline, at week 13, and then every 3 months. Opioid analgesic dosage was continuously adjusted as clinically indicated. Toxicity was graded according to the Cancer and Leukemia Group B expanded common toxicity criteria. If grade 3 or 4 toxicity occurred, dosing was interrupted until the toxicity resolved to grade 2 or baseline. Patients who experienced any persistent (8 weeks or more) or recurrent grade 3 or 4 toxicity without significant antitumor response were considered to have reached DLT, and treatment was discontinued permanently.

Response Criteria
Pain and opioid analgesic use were the primary indicators of response. Each night, patients scored their worst pain over the prior 24 hours (the daily worst pain) from 0 (no pain) to 10 (pain as bad as you can imagine) as part of the Brief Pain Inventory. Each night, patients also recorded their daily opioid analgesic use, which was subsequently converted to morphine equivalents. Pain response was prospectively defined before unblinding of the data and was achieved when, for a minimum of 3 consecutive weeks, pain decreased three points from baseline while opioid analgesic use either decreased or remained stable ( < 15% increase), opioid analgesic use decreased 33% from baseline while pain either decreased or remained stable ( < 2 point increase), or both of these. For patients with baseline pain scores 2 but less than 3, a reduction to 0 was required to achieve response. Similarly, for patients with baseline daily opioid analgesic use 5 mg but less than 15 mg (morphine equivalents), a decrease by 5 mg was required to achieve pain response.

Prior reports^2-4 have suggested that posttherapy declines in PSA of 50% correlate with improved survival. For this reason, patients were retrospectively categorized into one of two categories on the basis of whether a greater than 50% decline from baseline that lasted 28 days was achieved (for purposes of this report, termed a PSA decline). Performance status was measured weekly with the Validated Revised Rand Functional Limitations Scale (RRFLS), which captured each patient’s self-assessment of everyday activities such as self-care and mobility, ranging from a score of 8 (least functional impairment) to 40 (most functional impairment), as well as by physician-determined Karnofsky Performance Status (KPS).

Disease Progression
Objective disease progression was defined as an increase in size of measurable lesions, development of new osseous lesions, new urinary outflow obstruction secondary to tumor that required intervention, new malignant pleural effusion, or new spinal cord compression. In all patients, regardless of their pain response status, demonstration of objective disease progression was considered evidence of treatment failure and resulted in unblinding of treatment. Pain (subjective) progression was prospectively defined as a two-point increase in weekly average of the daily worst pain score or a greater than 15% increase in weekly average of the total daily opioid analgesic intake, each with deterioration from baseline in RRFLS performance status by eight points. An accompanying decline in RRFLS score was mandated to make disease progression defined by pain, opioid requirements, or both more rigorous and therefore more likely to be clinically significant. Treatment assignment was unblinded when either disease progression or DLT occurred, and patients found to be on placebo were offered open-label suramin. A minimum of 6 weeks on study was required before treatment could be unblinded. Unblinding for disease progression before the 6-week mark required study chair approval.

Statistical Considerations
The objective progression-free survival (OPFS) time, time to pain progression (TTPP), and overall survival time were measured from the first day of treatment with the Kaplan-Meier method.¹² All analyses were performed on an intent-to-treat basis. Thus, patients who progressed on placebo plus hydrocortisone and who were crossed over to suramin plus hydrocortisone were included in the analysis.

To evaluate the association between PSA decline and other measures of outcome, the Kaplan-Meier method was used to compare, overall and within each treatment group, the trends for three end points (OPFS, TTPP, and overall survival) for patients with and without a PSA decline of 50%. All analyses were performed with the Landmark Method.¹³ Three different landmarks were used: week 6, week 9, and week 12. Patients whose survival was shorter than the landmark point were excluded from analysis. This method has been used as a means of reducing the inherent bias of assessing survival as a function of response.

The effect of other variables on the end points was evaluated by fitting a Cox model with that variable.¹⁴ The variables considered were two pretreatment stratification variables (PSA level 100 ng/mL v > 100 ng/mL and presence or absence of measurable disease) and size of treating center. (Study centers were assigned to one of three categories on the basis of accrual, with the goal of accounting for differing levels of familiarity and expertise with suramin administration that would naturally evolve with increasing enrollment. These arbitrary categories were [a] up to eight patients enrolled, [b] 9 to 23 patients enrolled, and [c] 24 or more patients enrolled.) In addition, age, race, baseline pain medication requirement, RRFLS performance status, KPS, alkaline phosphatase, hemoglobin, and lactate dehydrogenase values were considered. For each end point, the variable was first entered linearly and then (where applicable), entered as a group of tertiles. Models were fitted using the variable alone and the variable in the presence of the treatment group. Any variable with a P value of less than .10 was considered a possible confounder for the end point. Where the linear and the tertile were both significant, the linear form in the multivariate model was used. Once a list of candidate confounders was established, a Cox model was fitted with these confounders and PSA response.¹⁴ For the combined population (suramin plus hydrocortisone and placebo plus hydrocortisone), models were also fitted with and without a term for treatment group.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
From February 1994 to December 1996, 460 patients entered the study: 229 in the suramin plus hydrocortisone group and 231 in the placebo plus hydrocortisone group. All patients enrolled were included in an intent-to-treat analysis, with the exception of two patients (one on each arm) who were randomized but were found to be ineligible and did not receive treatment. The two groups of patients were balanced with regard to age, race, baseline pain score, baseline daily opioid analgesic requirements, RRFLS performance status, KPS, site of disease (bone only v bone plus soft tissue), PSA level, hemoglobin level, and prior hormonal therapy, as previously described. Of 230 patients randomized to placebo plus hydrocortisone, 164 (71.3%) crossed over to receive suramin after progressing.

Of 269 suramin patients, the percentage of patients with a 50% decrease in PSA lasting at least 28 days, was 23%, 39%, and 45% with a 6-week, 9-week, and 12-week landmark, respectively. The respective percentage of 231 patients receiving placebo with a greater than 50% PSA decline was 5%, 21%, and 30%. For all 460 patients enrolled, the percentage with a PSA decline at the three landmarks was 14%, 31%, and 38%. Duration of pain response, time to progression, and overall survival in the two treatment groups have been reported elsewhere.¹¹ Disease progression occurred in 376 (82%) of the 458 patients and was documented on the basis of objective evidence in approximately 85% of disease progressors. Subjective evidence of disease progression (an increase in pain, analgesic use, or both, combined with deterioration in RRFLS performance status) accounted for 12% and 17% of patients with progressive disease in the suramin and placebo groups, respectively. The details of patterns of disease progression have been reported elsewhere.¹¹

Association of PSA Decline With Overall Survival
When the entire cohort (suramin and placebo patients) is considered, a durable decline in PSA was found to be associated with improved median survival at the 6-week landmark (563 v 325 days, P = .0019), 9-week landmark (596 v 344 days, P = .0009), and 12-week landmark (597 v 374 days, P = .003). A decline in PSA in patients randomized to receive suramin was also associated with an increased median survival at the 6-week landmark (532 v 312 days, P = .0033), 9-week landmark (532 v 333 days, P = .0028), and 12-week landmark (596 v 392 days, P = .003). In the placebo plus hydrocortisone–treated patients, there was a trend toward increased survival in patients exhibiting a decline in PSA (600 v 337 days at the 6-week landmark, 612 v 355 days at the 9-week landmark, and 601 v 368 days at the 12-week landmark). However, these trends did not reach statistical significance, perhaps because of the small number of patients with a PSA decline in the placebo-treated patients ( Table 1).

View this table: [in this window] [in a new window]   Table 1. Association Between PSA Decline ( 50% decrease in PSA lasting at least 28 days) at Landmarks and Overall Survival Showing the P Value From Log-Rank Test (univariate analysis)

View larger version (15K): [in this window] [in a new window]   Fig 1. Overall survival (days) by PSA decline at the 9-week landmark (all patients). Median (days): response = 596 days; no response = 344 days (P = .0009).

View larger version (15K): [in this window] [in a new window]   Fig 2. Time to objective progression by PSA decline at 9-week landmark (all patients). Median (days): response = 183 days; no response = 96 days ( P < .0001.

View larger version (15K): [in this window] [in a new window]   Fig 3. Time to pain progression by PSA decline at 9-week landmark (all patients). Median (days): response = 428; no response = 184 ( P = .0009).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

In assessing the predictive utility of posttreatment PSA changes, previous reports have appropriately focused on the hardest end point: survival.^2-5 However, correlation of PSA declines with other markers of clinical benefit might strengthen the argument that a PSA decline can serve as an intermediate end point in clinical trials involving HRPC patients. In addition to analyzing the relationship between PSA decline and survival, this data set permitted a correlation with other clinically important end points, including median TTPP and OPFS. Criteria for both pain response and pain progression were prospectively defined before unblinding and well before any correlation with PSA data was undertaken. A significant association between PSA decline and both OPFS and TTPP was observed. This associationbetween PSA decline and clinically significant subjective end points such as pain control suggests that PSA changes may have greater ramifications than generally appreciated. To our knowledge, this study represents the largest prospective series reported correlating PSA changes with patient-derived end points such as pain control. Nevertheless, because the overall results of this study did not demonstrate a survival advantage of one treatment arm over the other, the conditions for surrogacy (of PSA vis-à-vis survival) were not met.²⁰

Several groups have previously reported on the predictive value of posttreatment decline in PSA in phase II trials for the treatment of HRPC.^2-5 Such analyses derived from uncontrolled experience should be viewed as hypothesis generating and require further testing in prospectively designed randomized trials. Investigators from Memorial Sloan-Kettering^2-4 and from the University of Michigan³ have shown that a durable greater than 50% decline in PSA at an 8-week landmark was an independent predictor of survival in HRPC patients. Investigators at the University of Maryland conducted a similar analysis on data collected on 103 patients enrolled onto two phase I and II studies of suramin. Although a relationship between PSA declines at 4 weeks and survival was seen, these investigators were unable to identify a specific threshold of PSA decline consistently associated with a survival advantage.⁵ This study may have failed to show such an association because of smaller sample size. In addition, this series included a phase I and II study, which included different dosing regimens that could conceivably have had different therapeutic effects.

Laboratory studies have suggested that suramin may affect PSA levels without necessarily resulting in cell death. Prostate cancer cell lines grown in xenograft models have been reported to have reduced expression of PSA mRNA without associated cell death when treated in vitro with suramin.¹⁰ This observation has been the basis of recommendations urging caution in the use of PSA as a predictor of response in suramin-treated patients.²¹ However, this data set demonstrates a significant relationship between PSA decline and clinically relevant outcomes (overall survival, time to objective progression, and TTPP) in suramin-treated patients. The results of this analysis suggest that preclinical assays designed to evaluate the effects of drugs on PSA secretion may not necessarily be clinically significant and require proper validation before they can be used to interpret results from clinical trials.

In summary, this database, derived from the treatment of symptomatic HRPC patients with either suramin plus hydrocortisone or placebo plus hydrocortisone on a prospective, placebo-controlled, double-blind trial, supports the observation that a durable decline in PSA is associated with overall survival. Furthermore, a durable PSA decline was also associated with other markers of response, including TTPP and OPFS. This analysis indicates that previously reported data on the in vitro effects of suramin on PSA secretion are likely to have limited clinical relevance and raises a larger question as to the utility of such preclinical assays.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Scher HI, Mazumdar M, Kelly WK: Clinical trials in relapsed prostate cancer: Defining the target. J Natl Cancer Inst 88: 1623-1634, 1996[Abstract/Free Full Text]

2. Kelly WK, Scher HI, Mazumdar M: Prostate-specific antigen as a measure of disease outcome in metastatic hormone-refractory prostate cancer. J Clin Oncol 11: 607-615, 1993[Abstract]

3. Smith DC, Dunn RL, Strawderman MS, et al: Change in serum prostate-specific antigen as a marker of response to cytotoxic therapy for hormone-refractory prostate cancer. J Clin Oncol 16: 1835-1843, 1998[Abstract]

4. Scher HI, Kelly WK, Zhang ZF, et al: Post-therapy serum prostate-specific antigen level and survival in patients with androgen-independent prostate cancer. J Natl Cancer Inst 91: 244-251, 1999[Abstract/Free Full Text]

5. Sridhara R, Eisenberger M, Sinibaldi V, et al: Evaluation of prostate specific antigen as a surrogate marker for response of hormone refractory prostate cancer to suramin therapy. J Clin Oncol 13: 2944-2953, 1995[Abstract]

6. Dawson NA: Apples and oranges: Building a consensus for standardized eligibility criteria and end points in prostate cancer clinical trials. J Clin Oncol 16: 3398-3405, 1998[Abstract]

7. Tannock I, Gospodarowicz M, Meakin W, et al: Treatment of metastatic prostate cancer with low-dose prednisone: Evaluation of pain and quality of life as pragmatic indices of response. J Clin Oncol 7: 590-597, 1989[Abstract]

8. Scher HI, Kelly WK: Flutamide withdrawal syndrome: Its impact on clinical trials in hormone-refractory prostate cancer. J Clin Oncol 11: 1566-1572, 1993[Abstract/Free Full Text]

9. Small EJ, Srinivas S: The antiandrogen withdrawal syndrome: Experience in a large cohort of unselected advanced prostate cancer patients. Cancer 76: 1428-1434, 1995[Medline]

10. Thalmann GN, Sikes RA, Chang S-M, et al: Suramin-induced decrease in prostate-specific antigen expression with no effect on tumor growth in the LNCaP model of human prostate cancer. J Natl Cancer Inst 88: 794-801, 1996

11. Small EJ, Meyer M, Marshall ME, et al: Suramin therapy for patients with symptomatic hormone refractory prostate cancer: Results of a randomized phase III trial comparing suramin plus hydrocortisone to placebo plus hydrocortisone. J Clin Oncol 18: 1440-1450, 2000[Abstract/Free Full Text]

12. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958

13. Anderson JR, Cain KC, Gelber RD: Analysis of survival by tumor response. J Clin Oncol 1: 710-719, 1989[Abstract]

14. Cox DR: Regression models and life-tables (with discussion). J R Stat Soc (B) 34: 187-220, 1972

15. Fossa S, Aaronson N, Newling D: Quality of life and treatment of hormone-resistant metastatic prostate cancer. Eur J Cancer 26: 1133-1136, 1990

16. Cleeland CS, Ryan KM: Pain assessment: Global use of the Brief Pain Inventory. Ann Acad Med Singapore 23: 129-138, 1994[Medline]

17. Kornblith AB, Hollis DR, Zuckerman E, et al: Effect of megestrol acetate on quality of life in a dose-response trial in women with advanced breast cancer. J Clin Oncol 11: 2081-2089, 1993[Abstract/Free Full Text]

18. Kantoff PW, Conaway M, Winer E, et al: Hydrocortisone with or without mitoxantrone in patients with hormone refractory prostate cancer: Results of the Cancer and Leukemia Group B 9182 Study. J Clin Oncol 17: 2506-2513, 1999[Abstract/Free Full Text]

19. Tannock IF, Osoba D, Stockler MR, et al: Chemotherapy with mitoxantrone plus prednisone or prednisone alone for symptomatic hormone-resistant prostate cancer: A Canadian randomized trial with palliative end points. J Clin Oncol 6: 1756-1764, 1996

20. Prentice RL: Surrogate endpoints in clinical trials: Definition and operational criteria. Stat Med 8: 431-440, 1989[Medline]

21. Eisenberger MA, Nelson WG: How much can we rely on the level of prostate-specific antigen as an end point for evaluation of clinical trials? A word of caution! J Natl Cancer Inst 88: 779-781, 1996

Submitted July 31, 2000; accepted November 15, 2000.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				G. Colloca, F. Checcaglini, and A. Venturino About sorafenib in castration-resistant prostate cancer Ann. Onc., October 1, 2008; 19(10): 1812 - 1813. [Full Text] [PDF]

				C. P. Tarassoff, P. M. Arlen, and J. L. Gulley Therapeutic vaccines for prostate cancer. Oncologist, May 1, 2006; 11(5): 451 - 462. [Abstract] [Full Text] [PDF]

				D. P. Petrylak, D. P. Ankerst, C. S. Jiang, C. M. Tangen, M. H. A. Hussain, P. N. Lara Jr., J. A. Jones, M. E. Taplin, P. A. Burch, M. Kohli, et al. Evaluation of prostate-specific antigen declines for surrogacy in patients treated on SWOG 99-16. J Natl Cancer Inst, April 19, 2006; 98(8): 516 - 521. [Abstract] [Full Text] [PDF]

				K. J. Pienta and D. C. Smith Advances in Prostate Cancer Chemotherapy: A New Era Begins CA Cancer J Clin, September 1, 2005; 55(5): 300 - 318. [Abstract] [Full Text] [PDF]

				E. J. Small, S. Halabi, N. A. Dawson, W. M. Stadler, B. I. Rini, J. Picus, P. Gable, F. M. Torti, E. Kaplan, and N. J. Vogelzang Antiandrogen Withdrawal Alone or in Combination With Ketoconazole in Androgen-Independent Prostate Cancer Patients: A Phase III Trial (CALGB 9583) J. Clin. Oncol., March 15, 2004; 22(6): 1025 - 1033. [Abstract] [Full Text] [PDF]

				O. Smaletz, H. I. Scher, E. J. Small, D. A. Verbel, A. McMillan, K. Regan, W. K. Kelly, and M. W. Kattan Nomogram for Overall Survival of Patients With Progressive Metastatic Prostate Cancer After Castration J. Clin. Oncol., October 1, 2002; 20(19): 3972 - 3982. [Abstract] [Full Text] [PDF]

				E. J. Small, S. Halabi, M. J. Ratain, G. Rosner, W. Stadler, D. Palchak, E. Marshall, R. Rago, V. Hars, G. Wilding, et al. Randomized Study of Three Different Doses of Suramin Administered With a Fixed Dosing Schedule in Patients With Advanced Prostate Cancer: Results of Intergroup 0159, Cancer and Leukemia Group B 9480 J. Clin. Oncol., August 15, 2002; 20(16): 3369 - 3375. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Small, E. J. Articles by Eisenberger, M. Search for Related Content PubMed PubMed Citation Articles by Small, E. J. Articles by Eisenberger, M. Social Bookmarking                            What's this?