Saphenous vein graft failure as a Surrogate Endpoint in Clinical Trials: Difference between revisions
No edit summary |
Sergekorjian (talk | contribs) No edit summary |
||
| Line 96: | Line 96: | ||
===Pathophysiologic Basis as to Why SVG Failure is Associated with Adverse Outcomes=== | ===Pathophysiologic Basis as to Why SVG Failure is Associated with Adverse Outcomes=== | ||
[[Image:Reasons why svg failure is bad.gif|thumb|600px| ]] | [[Image:Reasons why svg failure is bad.gif|thumb|600px|none]] | ||
===Association of Primary PCI of Acutely Occluded SVGs with Adverse Events=== | ===Association of Primary PCI of Acutely Occluded SVGs with Adverse Events=== | ||
Latest revision as of 12:25, 20 April 2016
Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1]
Overview
One issue that arises is the suitability of SVG patency, and SVG narrowing as a surrogate endpoint for clinical events. A surrogate marker (or surrogate end point) is term used in medical research for a change to the human body that is believe to be necessary to an eventual outcome or end point.[1] In clinical trials, a surrogate endpoint is a measure of the effect (often what is known as a biomarker like cholesterol) of a certain treatment that is substituted in the evaluation of a new drug or device in place of a "hard endpoint" like death or heart attack. The National Institutes of Health (USA) defines surrogate endpoints as: "A biomarker intended to substitute for a clinical endpoint".[2] The use of a surrogate endpoint can lead to more rapid and efficient completion of clinical trials, but the use of a surrogate endpoint has been criticized as reductions of a substitute maker are not always predictive of improvements in clinical outcomes. The classic example of a failed surrogate marker is the use of PVC suppression as a substitute marker of anti-arrhythmia effectiveness. Class III anti-arrhythmias reduced PVCs or extra heart beats, but were associated with a higher risk of death in the Cardiac Arrhythmia Suppression Trial (CAST)[3]. Surrogate markers may be used when it is unethical to look for the end point (e.g., death) in the experiment, or when the number of end point events is very small, thus making it impractical to conduct an experiment to look for the end point. The measurement of surrogate markers provides a way to test the effectiveness of a treatment for a fatal disease without having to wait for a statistically significant number of deaths to occur. The FDA will often accept evidence from clinical trials that show a benefit with respect to surrogate markers instead of hard clinical end points.
The Advantages of a Surrogate Endpoint
The assessment of "hard" primary clinical endpoints (such as death and heart attack) often requires large long-term clinical trials which can be quite expensive. The use of surrogate endpoints can allow trials to evaluate the efficacy of a new drug or device more rapidly, more efficiently and more inexpensively.
The Disadvantages of Surrogate Endpoints
There are several potential disadvantages of a surrogate endpoint.
1. The surrogate endpoint may intuitively be hypothesized to be related to a "hard endpoint" such as death or heart attack, but may not be.
2. While a surrogate endpoint may be related to a "hard endpoint" such as death or heart attack, it is not clear that a reduction in the surrogate endpoint will lead to an improvement in the "hard endpoint" in death or heart attack. Anti-diabetic agents have been shown to reduce long term glucose (Hemoglobin A1c or HbA1C). It was hypothesized that more intense glucose control (a reduction in HbA1C) would be associated with a lower rate of death and heart attack. However, despite lowering of HbA1C, rosiglitazone or Avandia was not associated with a reduction in death and MI, but with an increase in the RECORD trial [4] [5].
In another example, class III antiarrhythmic agents were associated with suppression of premature ventricular contractions or PVCs, but were associated with a higher rate of death.
3. While a surrogate endpoint may be related to a "hard endpoint, it may be an acausal association (the surrogate may not lie in the causal pathway to the "hard endpoint" and changing the surrogate endpoint may not change the "hard endpoint".)
4. The agent may reduce the surrogate endpoint, but due to off target toxicity, may increase the risk of "hard endpoints" such as death or MI. Lower HDL is associated with a higher risk of adverse cardiac outcomes, Torcetrapib raises HDL and should therefore improve clinical outcomes, however, Torcetrapib administration was found to be associated with a higher rate of adverse clinical outcomes. It was felt that the potential benefit of Torcetrapib was reversed due to off target toxicity of a slight increase in blood pressure associated with Torcetrapib administration.
5. The relationship between the surrogate endpoint and the "hard endpoint" may be non-linear or may be a threshold effect. For example, in antiplatelet agent studies, it is unclear if ever greater levels of inhibition of platelet aggregation are associated with ever greater reductions in adverse outcomes, or if one must achieve just a certain "threshold" level of inhibition to improve outcomes.
There have been a number of instances when studies using surrogate markers have been used to show benefit from a particular treatment, but later, a repeat study looking at endpoints has not shown a benefit, or even a harm.[6]
Examples of Surrogate Endpoints
Examples of surrogate markers include:
- Total cholesterol
- High density lipoprotein (HDL)
- Low density lipoprotein (LDL)
- c reactive protein (cRP)
- Platelet inhibition by light transmittance aggregometry
- Coronary blood flow
- Minimum diameter or an artery and restenosis
- Fragmented blood cells are a surrogate marker for organ failure or stroke in TTP
- The S-phase duration, may be used as a surrogate marker for breast cancer occurrence
- CD4 count is a surrogate marker for death from HIV infection
Evaluation of a Surrogate Endpoint
A surrogate may be suitable as a surrogate endpoint if the following three criteria are satisfied:
- The drug or device improves the surrogate
- Improvement in the surrogate is related to an improvement in a hard clinical endpoint
- The same drug or device improves the hard clinical endpoint
In order for a surrogate to be validated, there must be at least one large clinical trial that satisfies these criteria. After this, other studies could rely upon the surrogate if it has been validated in a large study.
Caveats in the use of SVG Failure as a Surrogate Endpoint
There are multiple studies demonstrating that SVG closure or failure is associated with clinical events. The association of SVG failure or narrowing with clinical events is complex and should be interpreted in light of the following nuances:
- Often a composite endpoint of death, MI and revascularization is reported. One must be careful to deconstruct this composite endpoint to look at the association between SVG failure and each of these endpoints separately.
- It can be useful to distinguish between an MI that precedes and MI that follows revascularization to treat SVG failure. Understanding the timing of the myonecrosis allows one to undersantd the mechanism that caused the MI. An MI that precedes revascularization may be temporally attributable to the SVG failure itself, rather than the revascularization procedure to treat the SVG failure. A limitation to this assertion is that the release of markers of myonecrosis may be delayed, and a late rise in a cardiac biomarker may not be solely due to the revascularization, but it may also be related to the original occlusion of the SVG. After PCI for early SVG failure, 20.6% sustain an MI, and 30.6% of PCIs for late SVG failure are associated with MI (creatine kinase-myocardial band (CK-MB) greater than twice normal) [7]. However, whether the MI preceded the intervention to treat SVG failure, or followed the intervention to treat SVG failure, SVG failure can reasonably be deemed to be the proximate cause of the event.
- One must be careful to distinguish between a) immediate peri-operative MIs during the index hospitalization, b) early MIs between the index hospitalization and a year, and c) late MIs (after a year). If one is evaluating a therapy targeted at reducing late atherothrombosis or drugs that reduce intimal hyperplasia, then events between b and c are most relevant. If one is investigating a therapy targeting thrombotic occlusion, then all periods (a,b, and c) are relevant.
- In observational studies, those patients who are alive to undergo repeat angiography are obviously included in the analyses, and those patients who died prior to angiography may not be included in the observational analysis (i.e. there may not be imputation of an SVG failure to those patients who died). This results in underestimation of the association between SVG failure and mortality. For example, a report from the Duke Databank included those patients who survived to undergo subsequent cardiac catheterization, and excluded those patients who died by did not undergo cardiac catheterization [8].
- The SVG may close and the native artery may remain open minimizing the symptoms associated with SVG failure. SVG failure may therefore by clinically "silent" or may not be closely associated with "hard endpoints" such as myocardial infarction(MI).
- Despite the native artery remaining open after SVG failure, it should be realized that placement of an SVG is associated with more rapid disease progression in the grafted native coronary artery. SVG placement therefore exposes the patient to a risk of more rapid native disease progression. There is a 4 to 6 fold increase in the risk of proximal native vessel occlusion following the placement of an SVG, while there is limited impact upon disease progression downstream from the SVG [9][10][11][12] . While long-term studies demonstrate that as many as 22/23 grafted vessels occlude proximal to the SVG insertion site, the patency beyond the SVG insertion remains better (only 8 of 39 segments failed)[12]. Progression of native disease is more rapid in segments bypassed by and SVG than those bypassed by an arterial conduit (p = 0.001, odds ratio = 2.03)[13]. In summary, if the SVG fails, the patient most often is left with greater progression of the underlying native vessel disease than they would have had had they not had an SVG placed. It is difficult to ascertain the impact of native vessel disease acceleration given the limited duration of follow-up in trials of SVGs. The impact of disease progression may not be apparent for many years and may be underestimated in current trials and analyses.
- As a nuance of the above, the impact of SVG failure may depend in part upon when it occurs.
- SVG failure not only leads to a potential reduction in antegrade blood flow to the bypassed segment as a result of vessel closure, there is also the potential for embolizaiton from a large occluded conduit into the downstream native circulation. SVGs do not have sidebranches and there is therefore no capacity for alternate run off when occlusion occurs. As a result, SVGs often occlude back to the origin of the SVG. Furthermore, the diameter of SVG often exceeds that of native arteries. As a result of the fact that the SVG occludes back to the ostium and is of a larger volume than a native coronary artery, there is a much larger embolic burden associated with a SVG.
- Consistent with the embolic hazard cited above, is the fact that dilation of an SVG failure does not lead to improved clinical outcomes. The fact that opening a closed SVG does not lead to improved outcomes may lead to inappropriate confusion surrounding SVG failure as a relevant clinical outcome. While it may be intuitive that opening an occluded or failed SVG would lead to improved outcomes (consistent with SVG failure being a surrogate), opening the occluded SVG may instead lead to embolizaiton of a large amount of the thrombotic material downstream. Preventing SVG occlusion may be related to improved outcomes, but treating SVG occlusion after it occurs may not be related to improved outcomes. The failure of reopening an occluded SVG to improve outcomes should not detract from the importance and relevance of preventing SVG occlusion in the first place as a valid surrogate endpoint.
- The patient is the unit of randomization, but failure of an SVG is the unit most closely related to clinical events.
- The impact of SVG failure must be interpreted in the context of the presence and patency of arterial conduits. The impact of SVG failure that includes failure of an SVG to the left anterior descending may be quite different than failure of an SVG to the RCA in the presence of a patent left internal mammary artery.
- The impact of SVG failure must be interpreted in the context of the patency and disease status of the native vessel.
- The impact of SVG failure must be interpreted in the context of left ventricular function as well as associated co-morbidities (renal failure, diabetes and advanced age which are all associated with SVG failure).
- Is the timing of SVG failure relevant? If an anticoagulant (either antiplatelet or antithrombin) is undergoing evaluation of its efficacy in the prevention of thrombotic graft closure, then it is irrelevant if this thrombotic closure is early (in the immediate peri-operative period) or late. It could be hypothesized that the clinical benefit of the anticoagulant agent would be operative during both the early and late follow-up periods.
- Patients may not return for follow-up angiography and this may result in ascertainment bias. Maybe the patient is feeling so well they don't feel compelled to return for repeat angiography (you missed a positive treatment effect), or maybe they are so sick they can't show up (you missed a negative treatment effect). If patients died, traditionally they are counted or imputed as having SVG failure on both a per patient and a per SVG conduit basis. The percent of patients who returned for follow-up angiography was 80% in the PREVENT 4 study [14]
- The patient may return for follow-up angiography, but one or more of the SVGs cannot be engaged and selectively injected. As stated above, in the PREVENT 4 study, the use of saphenous vein graft markers improved the odds of finding SVGs in particular occluded SVGs [15]. However, the SVG stenosis 70% or greater at follow-up did not differ by use of markers (25.8% with marker vs 24.4% without marker, p = not significant). In other words, there does not seem to be ascertainment bias whereby failure to find the SVG results in a different outcome (the outcomes in patients with SVG markers with a greater degree of ascertainment of the endpoint were no different).
- There are both per patient and per SVG units of analysis. Analyses should be presented on both a per patient basis (the unit of randomization) and a per SVG basis (the unit that is associated with clinical events). Because the behavior of multiple SVGs may be correlated, and this within patient correlation may reduce the estimate of the variance in the population, an adjustment for the within patient correlation must be provided when presenting the results on a per SVG basis[16]. The within patient conocrdance can be adjusted for using a General Linear Model of Intraclass Correlation (GLIMIC) [16].
Association of SVG Failure with Clinical Events
There are multiple mechanisms whereby SVG patency is related to clinical outcomes:
- Closure of the conduit may reduce antegrade blood flow
- There may be embolization into the native vessel from the thrombosed SVG conduit, the diameter of which often exceeds the native coronary artery
- The SVG may have accelerate native vessel disease which predisposes the patient to adverse outcomes when the vessel occludes
Fitzgibbon and colleagues evaluated the clinical outcomes among 1,388 patients with 5,065 grafts over 25 years at a single center. SVG occlusion occurring 1 year after CABG was associated with reoperation and mortality [17].
In the PREVENT 4 study, the rate of death or MI outside the immediate perioperative period was 13.9% among patients with SVG failure and 0.9% among those without SVG failure [14]. The rate of immediate perioperative MI among patients with SVG failure was 13.4 and 6.8% among those patients without SVG failure.
Association of SVG Failure after Stenting with Clinical Events
The impact of SVG failure on clinical outcomes after SVG stenting was demonstrated among 80 patients with 112 lesions in the SOS (Stenting Of Saphenous Vein Grafts) Trial [18]. Compared with BMS, DES (specifically paclitaxel elution) was associated with a reduced risk of SVG failure. SVG failure after stenting was associated with an acute coronary syndrome in 10 of the 24 patients (42%) . 7 of the 24 (29%) of patients presented with a non–ST-segment elevation acute myocardial infarction). Stable angina was present in 9 (37%) of SVG failure patients, and there were no symptoms in 5 (21%) SVG failure patients. The authors concluded that SVG failure after stenting presents as an acute myocardial infarction in 7/24 (30%) of cases. It is quite clear that SVG failure in an SVG that has had a stent placed is associated with worse clinical outcomes, and by extension, it stands to reason that SVG failure may be associated with worse clinical outcomes as well.
Pathophysiologic Basis as to Why SVG Failure is Associated with Adverse Outcomes
Association of Primary PCI of Acutely Occluded SVGs with Adverse Events
In several studies, if an SVG is the culprit vessel in acute MI, mortality is quite high [19].
Indeed, some investigators have demonstrated that the mortality associated with SVG occlusion presenting as an MI and requiring intervention may be greater than that of native vessels [20]. In the Mayo experience, a total of 1,072 patients with acute MI underwent primary PCI without prior lytics between 1991 and 1997. A total of 128 patients had previously undergone CABG and 944 had not undergone CABG. Of the patients who had previously undergone CABG, the primary PCI was performed in the native vessel in 65, and in the SVG itself in 63. In a multivariate model adjusting for co-morbidities, primary PCI of an SVG was independently associated with adverse cardiac events (death/MI/Repeat revascularization) (relative risk 1.48 [95% confidence interval 1.07-2.03], P =.02), but a history of prior CABG itself was not (relative risk 1.22 [95% confidence interval 0.96-1.56], P =.11).
In another study, patients with a thombosed SVG were at very high risk of subsequent events including mortality. [21]The acute and long-term outcomes among 34 consecutive patients who underwent PCI of 36 acutely occluded SVGs between 2003 and 2009 at one institution were evaluated. Although PCI of the SVG was successful in 81% of the patients, 39% of these patients sustained stent thrombosis. After a mean follow-up of 2.3 ± 1.9 years, the mortality at 1 year was 8% and the mortality at 3 years was 42%. An acute coronary syndrome occurred in 15% of patients at 1 year and in 41% of patients at 3 years.
References
- ↑ Cohn JN (2004). "Introduction to Surrogate Markers". Circulation. American Heart Association. 109: IV20&ndash, 1. PMID 15226247. Retrieved 2007-01-10.
- ↑ Controlled Clinical Trials 22:485–502 (2001))
- ↑ Naccarelli GV, Dougherty AH, Wolbrette D, Wiggins S (1991). "A critical appraisal of the cardiac arrhythmia suppression trial (CAST)". Applied Cardiopulmonary Pathophysiology : ACP. 4 (1): 9–16. PMID 10147539.
|access-date=requires|url=(help) - ↑ Home PD, Pocock SJ, Beck-Nielsen H, Curtis PS, Gomis R, Hanefeld M, Jones NP, Komajda M, McMurray JJ (2009). "Rosiglitazone evaluated for cardiovascular outcomes in oral agent combination therapy for type 2 diabetes (RECORD): a multicentre, randomised, open-label trial". Lancet. 373 (9681): 2125–35. doi:10.1016/S0140-6736(09)60953-3. PMID 19501900. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ Nissen SE (2010). "Setting the RECORD Straight". JAMA : the Journal of the American Medical Association. 303 (12): 1194–5. doi:10.1001/jama.2010.333. PMID 20332408. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ Psaty BM, Weiss NS, Furberg CD; et al. (1999). "Surrogate end points, health outcomes, and the drug approval process for the treatment of risk factors for cardiovascular disease". JAMA. 282: 786&ndash, 790.
- ↑ Caños DA, Mintz GS, Berzingi CO, Apple S, Kotani J, Pichard AD, Satler LF, Suddath WO, Waksman R, Lindsay J, Weissman NJ (2004). "Clinical, angiographic, and intravascular ultrasound characteristics of early saphenous vein graft failure". Journal of the American College of Cardiology. 44 (1): 53–6. doi:10.1016/j.jacc.2004.03.045. PMID 15234406. Unknown parameter
|month=ignored (help) - ↑ Halabi AR, Alexander JH, Shaw LK, Lorenz TJ, Liao L, Kong DF, Milano CA, Harrington RA, Smith PK (2005). "Relation of early saphenous vein graft failure to outcomes following coronary artery bypass surgery". The American Journal of Cardiology. 96 (9): 1254–9. doi:10.1016/j.amjcard.2005.06.067. PMID 16253593. Unknown parameter
|month=ignored (help) - ↑ Borowski A, Vchivkov I, Ghodsizad A, Gams E (2008). "Coronary artery disease progression in patients who need repeat surgical revascularisation: the surgeon's point of view". Journal of Cardiovascular Medicine (Hagerstown, Md.). 9 (1): 85–8. doi:10.2459/JCM.0b013e328011439e. PMID 18268427. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ Hamada Y, Kawachi K, Yamamoto T, Nakata T, Kashu Y, Watanabe Y, Sato M (2001). "Effect of coronary artery bypass grafting on native coronary artery stenosis. Comparison of internal thoracic artery and saphenous vein grafts". The Journal of Cardiovascular Surgery. 42 (2): 159–64. PMID 11292927. Unknown parameter
|month=ignored (help);|access-date=requires|url=(help) - ↑ Rupprecht HJ, Hamm C, Ischinger T, Dietz U, Reimers J, Meyer J (1996). "Angiographic follow-up results of a randomized study on angioplasty versus bypass surgery (GABI trial). GABI Study Group". European Heart Journal. 17 (8): 1192–8. PMID 8869860. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ 12.0 12.1 Guthaner DF, Robert EW, Alderman EL, Wexler L (1979). "Long-term serial angiographic studies after coronary artery bypass surgery". Circulation. 60 (2): 250–9. PMID 312703. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ Manninen HI, Jaakkola P, Suhonen M, Rehnberg S, Vuorenniemi R, Matsi PJ (1998). "Angiographic predictors of graft patency and disease progression after coronary artery bypass grafting with arterial and venous grafts". The Annals of Thoracic Surgery. 66 (4): 1289–94. PMID 9800822. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ 14.0 14.1 Alexander JH, Hafley G, Harrington RA, Peterson ED, Ferguson TB, Lorenz TJ, Goyal A, Gibson M, Mack MJ, Gennevois D, Califf RM, Kouchoukos NT (2005). "Efficacy and safety of edifoligide, an E2F transcription factor decoy, for prevention of vein graft failure following coronary artery bypass graft surgery: PREVENT IV: a randomized controlled trial". JAMA : the Journal of the American Medical Association. 294 (19): 2446–54. doi:10.1001/jama.294.19.2446. PMID 16287955. Unknown parameter
|month=ignored (help);|access-date=requires|url=(help) - ↑ Olenchock SA, Karmpaliotis D, Gibson WJ, Murphy SA, Southard MC, Ciaglo L, Buros J, Mack MJ, Alexander JH, Harrington RA, Califf RM, Kouchoukos NT, Ferguson TB, Gibson CM (2008). "Impact of saphenous vein graft radiographic markers on clinical events and angiographic parameters". The Annals of Thoracic Surgery. 85 (2): 520–4. doi:10.1016/j.athoracsur.2007.10.061. PMID 18222256. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ 16.0 16.1 Gibson CM, Kuntz RE, Nobuyoshi M, Rosner B, Baim DS (1993). "Lesion-to-lesion independence of restenosis after treatment by conventional angioplasty, stenting, or directional atherectomy. Validation of lesion-based restenosis analysis". Circulation. 87 (4): 1123–9. PMID 8462141. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ Fitzgibbon GM, Kafka HP, Leach AJ, Keon WJ, Hooper GD, Burton JR (1996). "Coronary bypass graft fate and patient outcome: angiographic follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during 25 years". Journal of the American College of Cardiology. 28 (3): 616–26. PMID 8772748. Unknown parameter
|month=ignored (help) - ↑ Lichtenwalter C, de Lemos JA, Roesle M, Obel O, Holper EM, Haagen D, Saeed B, Iturbe JM, Shunk K, Bissett JK, Sachdeva R, Voudris VV, Karyofillis P, Kar B, Rossen J, Fasseas P, Berger P, Banerjee S, Brilakis ES (2009). "Clinical presentation and angiographic characteristics of saphenous vein graft failure after stenting: insights from the SOS (stenting of saphenous vein grafts) trial". JACC. Cardiovascular Interventions. 2 (9): 855–60. doi:10.1016/j.jcin.2009.06.014. PMID 19778774. Unknown parameter
|month=ignored (help) - ↑ Nguyen TT, O'Neill WW, Grines CL, Stone GW, Brodie BR, Cox DA, Grines LL, Boura JA, Dixon SR (2003). "One-year survival in patients with acute myocardial infarction and a saphenous vein graft culprit treated with primary angioplasty". The American Journal of Cardiology. 91 (10): 1250–4. PMID 12745114. Unknown parameter
|month=ignored (help) - ↑ Al Suwaidi J, Velianou JL, Berger PB, Mathew V, Garratt KN, Reeder GS, Grill DE, Holmes DR (2001). "Primary percutaneous coronary interventions in patients with acute myocardial infarction and prior coronary artery bypass grafting". American Heart Journal. 142 (3): 452–9. PMID 11526358. Unknown parameter
|month=ignored (help) - ↑ Abdul-rahman R. Abdel-karim, MD; Subhash Banerjee, MD; Emmanouil S. Brilakis, MD, PhD. Percutaneous Intervention of Acutely Occluded Saphenous Vein Grafts: Contemporary Techniques and Outcomes. J Invasive Cardiol. 2010;22(6):253-257.