Advertisement
Mayo Clinic Proceedings Home

Hard-Wired Bias

How Even Double-Blind, Randomized Controlled Trials Can Be Skewed From the Start
  • Vinay Prasad
    Correspondence
    Correspondence: Address to Vinay Prasad, MD, MPH, Medical Oncology Branch, National Cancer Institute, National Institutes of Health, 10 Center Dr 10/12N226, Bethesda, MD 20892.
    Affiliations
    Medical Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD
    Search for articles by this author
  • Vance W. Berger
    Affiliations
    University of Maryland Baltimore County, Biometry Research Group, Division of Cancer Prevention, National Cancer Institute, National Institutes of Health, Rockville, MD
    Search for articles by this author
      Well-designed, adequately powered randomized controlled trials (RCTs) are rightfully considered the highest form of evidence on which to base treatment and diagnostic decisions, minimizing potential biases, particularly confounding, that plague nonrandomized evidence.
      Evidence-Based Medicine Working Group
      Evidence-based medicine. A new approach to teaching the practice of medicine.
      • Ioannidis J.P.
      Why most published research findings are false.
      • Prasad V.
      • Jorgenson J.
      • Ioannidis J.P.
      • Cifu A.
      Observational studies often make clinical practice recommendations: an empirical evaluation of authors’ attitudes.
      At the same time, simply using an RCT is not sufficient to ensure that conclusions are free from bias. In recent years, sponsors and trialists have incorporated subtle design choices into RCTs that have skewed the final trial results. Here, we describe the phenomenon of “hard-wired” bias—bias that is introduced at the outset of randomized trials.
      We hope that recognition of hard-wired bias serves as a reminder that there is room for improvement in the analysis and conduct of clinical trials.
      Bias in the interpretation of clinical data may occur at many junctures, as depicted in, but not limited to, the Figure. Selective reporting of outcomes and publication bias occur at the final step in the presentation of data. Bias in the interpretation and analysis of data, including deviations from preplanned statistical analysis, occur penultimately. Bias can also occur before the collection of data—implicit in the design of clinical trials. Unlike bias in the analysis or reporting of data, bias in the trial design cannot be corrected by using statistical methods or reanalysis
      • Ebrahim S.
      • Sohani Z.N.
      • Montoya L.
      • et al.
      Reanalyses of randomized clinical trial data.
      ; instead it can only be noted as a limitation of the study. What are some of the ways in which bias in clinical trials can become hard-wired?
      Figure thumbnail gr1
      FigureOrigins of bias in clinical trials.

      Crossover

      Crossover from placebo or control to the investigational agent is typically done in pharmacokinetic studies, or for interventions aimed at assessing a subjective end point (permitting intrauser comparisons). However, in modern clinical trials, crossover is increasingly used in studies testing the basic efficacy of a novel compound. For instance, many RCTs of cancer treatments allow patients assigned to placebo to crossover to treatment upon progression of their disease.
      • Prasad V.
      • Grady C.
      The misguided ethics of crossover trials.
      Such crossover affects interpretations of the drug’s effect on survival. For instance, if a cancer drug delays progression, but does not improve survival, crossover is often cited as the reason for these findings. The drug would have improved survival had it not been for crossover. However, despite this common interpretation, there are other valid interpretations in this setting. For instance, a drug may slow progression, but increase off-target deaths, such that it has no net benefit.
      • Prasad V.
      Double-crossed: why crossover in clinical trials may be distorting medical science.
      In this case, crossover can mask the harms of the medication and provide a misleading inference about benefits.
      Sipuleucel-T (Provenge, Dendreon) is a cancer vaccine approved by the US Food and Drug Administration for the treatment of metastatic prostate cancer. In the seminal trial leading to approval, the drug managed to improve overall survival without any evidence that it slowed disease progression (progression free survival was no different between arms).
      • Di Lorenzo G.
      • Ferro M.
      • Buonerba C.
      Sipuleucel-T (Provenge®) for castration-resistant prostate cancer.
      In the trial, many patients in the control arm received the frozen vaccine when their cancer progressed and fewer patients received docetaxel, a drug with a proven survival benefit, or received it after a delay. This design led to the suggestion that Sipuleucel-T exhibited efficacy not by improving outcomes but rather because crossover harmed the control group by delaying alternate effective therapy.

      Outcomes of Sipuleucel-T Therapy. Centers for Medicare & Medicaid Services Web site. http://www.cms.gov/Medicare/Coverage/DeterminationProcess/downloads/id77TA.pdf. Accessed March 6, 2014.

      When it comes to crossover, better conclusions cannot be drawn simply from the data; all interpretations of the data have to make some assumptions about whether crossover is benefiting or harming the control group. Consider RECORD-1, a randomized trial comparing everolimus to placebo in patients with metastatic renal cell cancer for whom previous therapy had failed. RECORD-1
      • Motzer R.J.
      • Escudier B.
      • Oudard S.
      • et al.
      RECORD-1 Study Group
      Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial.
      reported an improvement in progression-free survival, but no change in overall survival. This deficiency was attributed to crossover,
      • Motzer R.J.
      • Escudier B.
      • Oudard S.
      • et al.
      RECORD-1 Study Group
      Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial.
      and the manufacturer provided modeling experiments, arguing what the survival would have been were it not for crossover.
      • Korhonen P.
      • Zuber E.
      • Branson M.
      • et al.
      Correcting overall survival for the impact of crossover via a rank-preserving structural failure time (RPSFT) model in the RECORD-1 trial of everolimus in metastatic renal-cell carcinoma.
      However, this exercise relies on assumptions, which may not be true. Because we cannot know the true effect of everolimus on survival, it was rejected by the UK’s National Institute for Clinical Excellence.

      Everolimus for the second-line treatment of advanced renal cell carcinoma. National Institute for Clinical Excellence Web site. http://www.nice.org.uk/guidance/TA219/chapter/4-Consideration-of-the-evidence. Published April 2011. Accessed August 30, 2014.

      For a different set of questions however, a lack of crossover can bias results. When clinical trials seek to establish the basic efficacy of a novel compound, the presence of crossover can distort inferences, as we have shown. However, when clinical trials test the proper sequencing of agents already known to confer benefit, the absence of crossover can be equally problematic. Consider some contemporary examples. In a randomized trial from Spain that reported that lenalidomide and dexamethasone improved overall survival in patients with smoldering myeloma over the current standard of care,
      • Mateos M.-V.
      • Hernández M.-T.
      • Giraldo P.
      • et al.
      Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma.
      patients in the control arm were not routinely given access to lenalidomide when they developed overt multiple myeloma. Thus, we cannot be sure that the survival advantage of early treatment would still exist if control patients had fair access to this drug, as they would have had in the United States. The failure to prescribe lenalidomide upon progression is also a limitation of randomized trials of maintenance therapy,
      • Attal M.
      • Lauwers-Cances V.
      • Marit G.
      • et al.
      IFM Investigators
      Lenalidomide maintenance after stem-cell transplantation for multiple myeloma.
      • Palumbo A.
      • Cavallo F.
      • Gay F.
      • et al.
      Autologous transplantation and maintenance therapy in multiple myeloma.
      which should ask the question of whether continuous administration of a drug is better than receiving the agent upon having progressive disease.
      In a final example, the proteasome inhibitor bortezomib, which is Food and Drug Administration approved for relapsed mantle cell lymphoma, was tested in the first-line setting of this disease. At the time of the trial, the drug was widely used in the relapsed setting. Thus, any trial seeking to advance the drug into the first-line setting should show that the early use of the drug improves survival beyond its second-line use. However, the randomized trial testing this question was globally conducted, and, as such, patients in the control arm had poor access to the drug in the second-line setting, with only 19% receiving it.
      • Robak T.
      • Huang H.
      • Jin J.
      • et al.
      (LYM-3002 Investigators)
      Bortezomib-based therapy for newly diagnosed mantle-cell lymphoma.
      Thus, we cannot be sure that the survival advantage seen in the trial would exist had it been conducted in nations in which bortezomib is a mainstay of second-line treatment.

      Selection Bias

      Bias in selecting study subjects is a frequent concern in clinical trials. Selection bias can be caused by the inclusion and exclusion criteria of a study, which prevents the generalizability of results to a broader patient population. For instance, in this issue of the Mayo Clinic Proceedings, Zimmerman and colleagues demonstrate that the inclusion criteria for randomized controlled trials of antidepressants have become increasingly narrow over time.
      • Zimmerman M.
      • Clark H.L.
      • Multach M.D.
      • Walsh E.
      • Rosenstein L.K.
      • Gazarian D.
      Have treatment studies of depression become even less generalizable? A review of the inclusion and exclusion criteria used in placebo-controlled antidepressant efficacy trials published during the past 20 years.
      The authors compared trials from 2010-2014 against those in the preceding decade, and found that modern studies were more likely to exclude comorbid Axis 1 and personality disorders, and required more severe depression according to validated scales. These design choices greatly impair the generalizability of these trials, which are certain to be used to justify these drugs in broader populations. Another example comes from the field of cancer medicine. Among cancer drugs approved by the US Food and Drug Administration between 1995 and 2002, demographic characteristics of patients were strikingly different from those of cancer patients in the United States.
      • Talarico L.
      • Chen G.
      • Pazdur R.
      Enrollment of elderly patients in clinical trials for cancer drug registration: a 7-year experience by the US Food and Drug Administration.
      Although the proportion of patients 65 years or older, 70 years or older, and 75 years or older was 60%, 46%, and 31%, respectively, among cancer patients in the United States, these age groups comprised only 36%, 20%, and 9% of patients in registration trials (P<.001). Fehrenbacher et al
      • Fehrenbacher L.
      • Ackerson L.
      • Somkin C.
      Randomized clinical trial eligibility rates for chemotherapy (CT) and antiangiogenic therapy (AAT) in a population-based cohort of newly diagnosed non-small cell lung cancer (NSCLC) patients.
      extended these findings and found that the inclusion criteria of contemporary randomized trials in non-small cell lung cancer would exclude most patients treated at Kaiser Permanente, a large insurer with a representative patient population.
      Although these examples are illustrative, this problem is relatively tractable, as care can be taken to prevent extrapolation to untested populations. However, in other cases, selection bias cannot be accounted for.
      Consider the open-label “run-in period.” The Heart Protection Study,
      Heart Protection Study Collaborative Group
      MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial.
      a randomized trial of 20,536 individuals with high cardiovascular risk, tested whether 40 mg/d of simvastatin could improve outcomes as compared with placebo. The trial found that the medication decreased major vascular events by 25%, and the authors go further, arguing that without nonadherence the improvement would have been 33%.
      Heart Protection Study Collaborative Group
      MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial.
      Notably, this trial used a 4-week placebo run-in period followed by a 4- to 6-week simvastatin run-in period before randomization. During this time, a patient’s primary doctor could remove a patient from randomization, and any patient could elect not to be randomized for “any reason.”
      • Armitage J.
      • Bowman L.
      • Collins R.
      • Parish S.
      • Tobert J.
      MRC/BHF Heart Protection Study Collaborative Group
      Effects of simvastatin 40 mg daily on muscle and liver adverse effects in a 5-year randomized placebo-controlled trial in 20,536 high-risk people.
      All together, 11,609 patients who were eligible for the study and began the run-in period dropped out before randomization.
      • Armitage J.
      • Bowman L.
      • Collins R.
      • Parish S.
      • Tobert J.
      MRC/BHF Heart Protection Study Collaborative Group
      Effects of simvastatin 40 mg daily on muscle and liver adverse effects in a 5-year randomized placebo-controlled trial in 20,536 high-risk people.
      Thus, more than a third of the patients who began the study were not randomized, and no set of specified inclusion criteria can define the set of patients who remained. Others have noted that the use of run-in periods can limit the applicability of study findings and can inflate estimates of benefits.
      • Pablos-Méndez A.
      • Barr R.G.
      • Shea S.
      Run-in periods in randomized trials: implications for the application of results in clinical practice.
      This occurs in part because run-in periods of the active drug test a different clinical question, whether discontinuation of a therapy is harmful, rather than whether initiating a therapy is beneficial.
      Open-label run-in periods were also problematic in the PARADIGM-HF trial.
      • McMurray J.J.
      • Packer M.
      • Desai A.S.
      • et al.
      PARADIGM-HF Investigators and Committees
      Angiotensin-neprilysin inhibition versus enalapril in heart failure.
      This study randomized 8442 patients predominantly with New York Heart Association class II and III heart failure to a combination of valsartan (an angiotensin receptor blocker) and sacubitril (the investigational agent) or enalapril (the control arm). Yet, before randomization, more than 10,500 patients entered into a run-in period. During this time, patients were sequentially treated with a median of 15 days of enalapril followed by 29 days of the combination medication. Nearly 20% of participants who began the study dropped out of the study during this time. Thus, the run-in period created an indefinable study population, namely, patients who met inclusion criteria and did not drop out after treated with a median of 15 days of enalapril followed by 29 days of the combination medication, as well as posed a different question, whether switching back to enalapril was better or worse than continuing the combination medication in such patients.

      The Unequal Playing Field

      Another way modern trials have hard-wired bias is by promoting an unequal comparison. For instance, a head-to-head trial of the tyrosine kinase inhibitors axitinib and sorafenib for the treatment of metastatic kidney cancer appears to be fair; however, a closer examination reveals problems.
      • Rini B.I.
      • Escudier B.
      • Tomczak P.
      • et al.
      Comparative effectiveness of axitinib versus sorafenib in advanced renal cell carcinoma (AXIS): a randomised phase 3 trial.
      Specifically, although the starting dose of both drugs is appropriate, the dose reductions for toxicity favor the axitinib arm. For similar adverse effects, sorafenib has steeper dose reductions, and for patients doing well on full-dose axitinib, the dose could even be increased. Collectively, this meant that axitinib was pushed to a higher dose and penalized less than sorafenib.
      • Prasad V.
      • Massey P.R.
      • Fojo T.
      Oral anticancer drugs: how limited dosing options and dose reductions may affect outcomes in comparative trials and efficacy in patients.
      It should be no surprise which medication was declared the winner of that study.
      Another type of unequal playing field occurs when 2 cancer drugs are tested head to head, but as a matter of fact more patients had already taken, and had the opportunity to have their cancer acquire resistance to, one of those drugs. In the ENDEAVOR randomized trial, patients with multiple myeloma who had had their disease progress upon treatment were randomized to bortezomib or carfilzomib in combination with dexamethasone.

      Carfilzomib doubles PFS over bortezomib in phase III multiple myeloma trial. OncLive Web site. http://www.onclive.com/web-exclusives/Carfilzomib-Doubles-PFS-Over-Bortezomib-in-Phase-III-Multiple-Myeloma-Trial. Published March 2, 2015. Accessed May 3, 2015.

      The results indicated a progression-free survival benefit for carfilzomib, but as a matter of fact, given the date these drugs were approved, most patients had the opportunity to be previously treated with bortezomib whereas few patients had the chance to previously receive carfilzomib. As a general rule in cancer, 2 drugs can be on average comparable, but each is less effective in patients already treated with that medication—a bias that the ENDEAVOR trial exploits.

      Control Arms

      The control for a clinical trial is ideally selected on the basis of the clinical question posed. Controls should reflect the best therapy currently being used in the target population, and for studies evaluating subjective end points, the control should be as close as possible to the investigational arm. If this condition is not met, a trial essentially uses a straw-man comparator. Many times, the use of a sham control has unmasked bias in studies supporting the use of a medical procedure. For instance, vertebroplasty, epidural steroid injection, and arthroscopic meniscectomy all required a sham controlled trial to report that the treatments had no benefit.
      • Redberg R.F.
      Sham controls in medical device trials.

      Censoring

      Informative censoring
      • Uno H.
      • Claggett B.
      • Tian L.
      • et al.
      Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis.
      in clinical trials can distort our perception of the benefits of a treatment. All survival analyses are based on the premise that censoring is uninformative—the patients censored are no different from those who are followed. However, this assumption should be questioned. In many trials for cancer treatment, censoring occurs because patients who cannot tolerate the study medication, or have excess toxicity, withdraw from the study. These patients are likely to be different from those who tolerate therapy well. In a recent study, Campigotto and Weller
      • Campigotto F.
      • Weller E.
      Impact of informative censoring on the Kaplan-Meier estimate of progression-free survival in phase II clinical trials.
      provide 2 examples in which patients who are censored are likely to have better or worse survival than those who are followed. The authors then provide a range of estimates for the outcome had these patients not been excluded, on the basis of simulation. However, if patients discontinue treatment, and are no longer followed, we cannot reconstruct their outcomes without making assumptions. We are left with a hard-wired bias.

      Conclusion

      RCTs remain the best way to draw sound conclusions about the efficacy and impact of drugs, devices, screening, and diagnostic tests, but unfortunately randomization does not ensure a fair trial. In this respect, randomization can be viewed as necessary but not sufficient for sound scientific decision making. The purpose of our analysis is not to disparage the growth of RCTs—which we believe is inevitable and unquestionably valuable—but to highlight persistent challenges. Many types of bias can be corrected by using individual patient level data, whereas other types of bias, so-called “hard-wired” bias, cannot be corrected after the fact.
      Medical trials involve the participation of human subjects who donate their time and energy to further the altruistic pursuit of improved medical care. As such, we have a moral obligation to ensure that research is capable of most honestly answering an important clinical question. The elements of the trial design that we discussed—crossover, drug run-in periods, use of inadequate controls, early censoring, selection bias, and duration of follow-up—are decisions made at the outset of a clinical trial and cannot be corrected later. We must work to remove hard-wired bias from clinical trials, and only time will tell if our existing system (in which trials are predominantly funded and conducted by industry) can meet this challenge.

      Acknowledgments

      The views and opinions of the authors do not reflect those of the National Cancer Institute.

      References

        • Evidence-Based Medicine Working Group
        Evidence-based medicine. A new approach to teaching the practice of medicine.
        JAMA. 1992; 268: 2420-2425
        • Ioannidis J.P.
        Why most published research findings are false.
        PLoS Med. 2005; 2: e124
        • Prasad V.
        • Jorgenson J.
        • Ioannidis J.P.
        • Cifu A.
        Observational studies often make clinical practice recommendations: an empirical evaluation of authors’ attitudes.
        J Clin Epidemiol. 2013; 66: 361-366.e4
      1. Cancer Prevention II (Recent Results in Cancer Research). No. 2. Springer, Berlin2008
        • Ebrahim S.
        • Sohani Z.N.
        • Montoya L.
        • et al.
        Reanalyses of randomized clinical trial data.
        JAMA. 2014; 312: 1024-1032
        • Prasad V.
        • Grady C.
        The misguided ethics of crossover trials.
        Contemp Clin Trials. 2014; 37: 167-169
        • Prasad V.
        Double-crossed: why crossover in clinical trials may be distorting medical science.
        J Natl Compr Canc Netw. 2013; 11: 625-627
        • Di Lorenzo G.
        • Ferro M.
        • Buonerba C.
        Sipuleucel-T (Provenge®) for castration-resistant prostate cancer.
        BJU Int. 2012; 110: E99-E104
      2. Outcomes of Sipuleucel-T Therapy. Centers for Medicare & Medicaid Services Web site. http://www.cms.gov/Medicare/Coverage/DeterminationProcess/downloads/id77TA.pdf. Accessed March 6, 2014.

        • Motzer R.J.
        • Escudier B.
        • Oudard S.
        • et al.
        • RECORD-1 Study Group
        Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial.
        Lancet. 2008; 372: 449-456
        • Korhonen P.
        • Zuber E.
        • Branson M.
        • et al.
        Correcting overall survival for the impact of crossover via a rank-preserving structural failure time (RPSFT) model in the RECORD-1 trial of everolimus in metastatic renal-cell carcinoma.
        J Biopharm Stat. 2012; 22: 1258-1271
      3. Everolimus for the second-line treatment of advanced renal cell carcinoma. National Institute for Clinical Excellence Web site. http://www.nice.org.uk/guidance/TA219/chapter/4-Consideration-of-the-evidence. Published April 2011. Accessed August 30, 2014.

        • Mateos M.-V.
        • Hernández M.-T.
        • Giraldo P.
        • et al.
        Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma.
        N Engl J Med. 2013; 369: 438-447
        • Attal M.
        • Lauwers-Cances V.
        • Marit G.
        • et al.
        • IFM Investigators
        Lenalidomide maintenance after stem-cell transplantation for multiple myeloma.
        N Engl J Med. 2012; 366: 1782-1791
        • Palumbo A.
        • Cavallo F.
        • Gay F.
        • et al.
        Autologous transplantation and maintenance therapy in multiple myeloma.
        N Engl J Med. 2014; 371: 895-905
        • Robak T.
        • Huang H.
        • Jin J.
        • et al.
        • (LYM-3002 Investigators)
        Bortezomib-based therapy for newly diagnosed mantle-cell lymphoma.
        N Engl J Med. 2015; 372: 944-953
        • Zimmerman M.
        • Clark H.L.
        • Multach M.D.
        • Walsh E.
        • Rosenstein L.K.
        • Gazarian D.
        Have treatment studies of depression become even less generalizable? A review of the inclusion and exclusion criteria used in placebo-controlled antidepressant efficacy trials published during the past 20 years.
        Mayo Clin Proc. 2015; 90: 1180-1186
        • Talarico L.
        • Chen G.
        • Pazdur R.
        Enrollment of elderly patients in clinical trials for cancer drug registration: a 7-year experience by the US Food and Drug Administration.
        J Clin Oncol. 2004; 22: 4626-4631
        • Fehrenbacher L.
        • Ackerson L.
        • Somkin C.
        Randomized clinical trial eligibility rates for chemotherapy (CT) and antiangiogenic therapy (AAT) in a population-based cohort of newly diagnosed non-small cell lung cancer (NSCLC) patients.
        J Clin Oncol. 2009; 27 ([abstract 6538]): 15s
        • Heart Protection Study Collaborative Group
        MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial.
        Lancet. 2002; 360: 7-22
        • Armitage J.
        • Bowman L.
        • Collins R.
        • Parish S.
        • Tobert J.
        • MRC/BHF Heart Protection Study Collaborative Group
        Effects of simvastatin 40 mg daily on muscle and liver adverse effects in a 5-year randomized placebo-controlled trial in 20,536 high-risk people.
        BMC Clin Pharmacol. 2009; 9: 6
        • Pablos-Méndez A.
        • Barr R.G.
        • Shea S.
        Run-in periods in randomized trials: implications for the application of results in clinical practice.
        JAMA. 1998; 279: 222-225
        • McMurray J.J.
        • Packer M.
        • Desai A.S.
        • et al.
        • PARADIGM-HF Investigators and Committees
        Angiotensin-neprilysin inhibition versus enalapril in heart failure.
        N Engl J Med. 2014; 371: 993-1004
        • Rini B.I.
        • Escudier B.
        • Tomczak P.
        • et al.
        Comparative effectiveness of axitinib versus sorafenib in advanced renal cell carcinoma (AXIS): a randomised phase 3 trial.
        Lancet. 2011; 378 ([published correction appears in Lancet. 2012;380(9856):1818]): 1931-1939
        • Prasad V.
        • Massey P.R.
        • Fojo T.
        Oral anticancer drugs: how limited dosing options and dose reductions may affect outcomes in comparative trials and efficacy in patients.
        J Clin Oncol. 2014; 32: 1620-1629
      4. Carfilzomib doubles PFS over bortezomib in phase III multiple myeloma trial. OncLive Web site. http://www.onclive.com/web-exclusives/Carfilzomib-Doubles-PFS-Over-Bortezomib-in-Phase-III-Multiple-Myeloma-Trial. Published March 2, 2015. Accessed May 3, 2015.

        • Redberg R.F.
        Sham controls in medical device trials.
        N Engl J Med. 2014; 371: 892-893
        • Uno H.
        • Claggett B.
        • Tian L.
        • et al.
        Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis.
        J Clin Oncol. 2014; 32: 2380-2385
        • Campigotto F.
        • Weller E.
        Impact of informative censoring on the Kaplan-Meier estimate of progression-free survival in phase II clinical trials.
        J Clin Oncol. 2014; 32: 3068-3074

      Linked Article