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 Weiss, R. B. Articles by Hudis, C. A. Search for Related Content PubMed PubMed Citation Articles by Weiss, R. B. Articles by Hudis, C. A.

An On-Site Audit of the South African Trial of High-Dose Chemotherapy for Metastatic Breast Cancer and Associated Publications

From the Georgetown University Medical Center, Washington, DC; Independent Consultant, Johannesburg, South Africa; and Breast Cancer Medicine Service, Memorial Sloan-Kettering Cancer Center, New York, NY.

Address reprint requests to Raymond B. Weiss, MD, 15304 Narcissus Way, Rockville, MD 20853-1744; email: rayweissmd{at}aol.com

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
PURPOSE: The randomized study reported by Bezwoda et al of high-dose chemotherapy (HDC) for treatment of metastatic breast cancer was audited on site to verify the study results. Additional published studies were reviewed to determine whether they had been subject to the required institutional oversight.

PATIENTS AND METHODS: Ninety patients were reported to have been randomized and treated on this trial. A log of the names, hospital numbers, entry dates, and regimen received had been provided by the principal investigator. A search of more than 15,000 sets of medical records available from two Johannesburg hospitals was performed to locate records for as many of these 90 patients as possible. Standard auditing techniques were used. Additional clinical trials published by Bezwoda were compared against the minutes of the University of the Witwatersrand Committee for Research on Human Subjects to verify review and approval.

RESULTS: Records for only 61 of the 90 patients could be found. Of these 61, only 27 had sufficient records to verify eligibility for the trial by the published criteria. Of these 27, 18 did not meet one or more eligibility criteria. Only 25 patients appeared to have received their assigned therapy temporally associated with their enrollment date, and all but three of these 25 received HDC. The treatment details of individual patients were at great variance from the published data. Nine other trials reported by Bezwoda were not reviewed or approved by the appropriate institutional committee despite statements to the contrary in the publications.

CONCLUSION: The multiple publications of this study do not report verifiable data, and nine other publications coauthored by the principal investigator contain at least one major untrue statement.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
HIGH-DOSE CHEMOTHERAPY (HDC) for treatment of various stages of breast cancer was developed during the 1980s. Phase II trials reporting results of such therapy in patients with both metastatic disease and early-stage, high-risk disease^1-4 suggested that the outcome for patients receiving this treatment was superior. At the beginning of the 1990s, prospective randomized trials comparing HDC to some form of conventional treatment were initiated in a number of countries, but they would take some years to accrue sufficient patients to address the scientific questions being asked.

The number of women undergoing such therapy soared in the United States during the early 1990s.^3,5 Editorials and articles have been published by both proponents and agnostics vigorously debating the value of HDC for breast cancer,^6-10 including two that appeared in 2000.^11,12 All such debates have involved the uncertainty of whether the therapy was truly advantageous or the result of patient selection bias.¹³ While the randomized trials were ongoing, there was no resolution of this debate.

In 1995, Bezwoda et al^14,15 published the results of what was the first completed randomized trial comparing HDC to some form of less intensive chemotherapy. Early results of this trial were presented at the 1992 annual meeting of the American Society of Clinical Oncology (ASCO),¹⁶ and Bezwoda recounted the results of this trial in six publications subsequent to 1995.^17-22 These results caused the number of patients treated with HDC to increase,¹² and they were then referenced and discussed in numerous other articles. As of February 2001, the 1995 article has been cited 354 times (Institute for Scientific Information, personal communication, March 2001). An editorial²³ accompanied the original article in the same journal issue. Although it was a trial of only 90 patients, its influence was greatly magnified by the intense debate ongoing in the United States about the value of HDC for breast cancer.

In January 2000, an audit of another randomized trial conducted by Bezwoda (for high-risk, primary breast cancer) was performed, and major irregularities in this study were discovered.²⁴ Concern was then raised²⁵ about the validity of the metastatic study published by Bezwoda. Officials of the involved institution desired a review of this other study and, through the South African Medical Research Council, invited us to do it. As part of this probe, we decided to review aspects of other publications of Bezwoda. We report the results of these investigations.

	METHODS AND SCOPE OF REVIEW

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
Although the metastatic disease study^14,15 was not part of the initial audit in January 2000,²⁴ Bezwoda voluntarily provided patient enrollment lists at that time for both of his reports of HDC for breast cancer. He also provided a copy of a protocol and consent form for each of these studies purporting to be the original documents used in the conduct of these two trials. The consent forms were alleged to have been used to obtain consent from the enrolled patients. Along with the two protocols, he also provided documents indicating the studies had been approved by the Pharmacy and Therapeutics Committee at the University of the Witwatersrand. These approvals had the same date and were purportedly signed by the chair of this institutional committee. However, no documents denoting approval by the University’s Committee for Research on Human Subjects (the institutional review board [IRB]) were provided. The enrollment list provided by Bezwoda for the metastatic disease study with patient names, hospital numbers, designated treatment, and date of registration formed the basis of this current audit. In addition, a MEDLINE search of all publications by Bezwoda et al back to 1987 was performed. Those articles reporting clinical trial results were assessed for statements regarding approval by the University’s IRB.

Using the enrollment list Bezwoda had provided, a search was initiated for the medical records of all 90 patients said to have been entered onto this trial. Patients had been treated at two hospitals (Hillbrow Hospital and Johannesburg Hospital), but the January 2000 auditors were informed by Bezwoda that most of the patients had been treated at Hillbrow. This hospital was closed in 1997, and after this closure, the oncology records were transferred to the Johannesburg Hospital and stored on a large trolley. This trolley contained approximately 3,000 sets of patient records, and they were systematically reviewed in their entirety. This search yielded a match for 23 records. Since this number was far short of the 90-patient total, the search was extended to the filing systems within the Johannesburg Hospital. This systems search included the oncology ward, the department of radiation oncology, and the hospital computer system (which also included microfiche files). No inpatient records from the Hillbrow Hospital had been transferred to microfiche or were available. The offices of the oncology research staff and Bezwoda were also searched, but no patient records were found there.

In the oncology ward, three files matching names and numbers on the enrollment list were found. In the department of radiation oncology, 45 patient records were identified, 15 of which were name-only matches and 30 of which were a match for a name, hospital number, or treatment regimen. The search of the Johannesburg Hospital computer system yielded 12 patient number and/or name matches. In total, in excess of 15,000 sets of patient records were meticulously searched to locate as many of the 90 names on the list as possible.

During the audit, care was taken to ensure that the patient file reviewed matched the patient named on Bezwoda’s list. In cases in which the name matched but the hospital number did not, the file was considered valid if there was a clear statement with respect to diagnosis and treatment regimen.

In order to establish what approval applications had been submitted to the Committee for Research on Human Subjects of the University of the Witwatersrand (the IRB) by Bezwoda, all minutes and agendas of the committee meetings were reviewed spanning January 1983 to August 1998. A MEDLINE search was performed to identify all publications from 1987 to 2000 in which Bezwoda was among the authors. Sixteen articles reporting clinical trial results in which the text stated explicitly that the study had undergone review and approval by this committee were compared with the committee’s records.

Bezwoda had two coauthors on the 1995 papers,^14,15 one of whom was also named on the 1992 ASCO abstract.¹⁶ Both individuals were invited to travel to Johannesburg and participate in the record review. Lesley Seymour, MB BCh, PhD, FCP, accepted this invitation; Roger Dansey, MD, declined. Bezwoda was not invited to participate.

The audit itself was conducted using the standard procedures developed in the cooperative group audits conducted in North America^26,27 using the protocol Bezwoda had provided and the details as published in his article.¹⁵ The auditors (along with Dr Seymour) reviewed all available items for each patient where something (even only several pages of files) had been located in the record search.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
The 1995 article¹⁵ states that "the study was performed according to the principles of the Declaration of Helsinki, [and] all patients entered onto the study gave informed consent and the study was approved by the Committee for Ethics on Human Experimentation [sic] of the University of Witwatersrand." The review of the minutes and agendas of the Committee for Research on Human Subjects, the correct name of the IRB, failed to substantiate that any study of HDC for breast cancer had ever been submitted for approval at this institution. Approval by this committee before initiation of any research on human subjects has been a requirement of the University and the South African Medical Research Council since well before 1990,^28,29 when this study was allegedly begun. In addition, in no record reviewed was there any consent form signed by a participant. Finally, Bezwoda admitted at his dismissal hearing in March 2000 that he had written the protocol for the adjuvant study just before the audit in January 2000, 9 years after he allegedly had begun the study.³⁰ The protocol for the metastatic study was identical in format and typescript to the adjuvant one; only the title and the conventional regimen used were different. In the metastatic protocol, the conventional regimen named was cyclophosphamide, mitoxantrone (Novantrone; Immunex Corp, Seattle, WA), and vincristine (CNV). In the adjuvant protocol, the regimen named was cyclophosphamide, doxorubicin (Adriamycin; Pharmacia & Upjohn Co, Bridgewater, NJ), and fluorouracil (CAF). However, Bezwoda admitted after the January 2000 audit that he had misrepresented the CAF regimen.²⁴ He also admitted that the approval document of the hospital’s Pharmacy and Therapeutics Committee for the adjuvant study was false.³⁰ The document for the metastatic study provided to the auditors was identical in format and text to the adjuvant study document and was actually dated (December 10, 1990) 3 weeks after the first patient entry date on the enrollment log. In addition, one of the Pharmacy and Therapeutics Committee reviewers named on this approval form did not even join the medical school’s staff until 1993, nearly 3 years after the alleged committee approval.

The 1995 article¹⁵ states that "randomization was performed by a random-number closed-envelope technique." Dr Seymour told the auditors that when she participated in this trial, Bezwoda kept a logbook in his possession that had the two regimens designated as "A" or "B" in random sequence in a long list of blank spaces. When a patient was to be entered, her name would be entered on the next line in this book, and the regimen to be administered would be designated by the accompanying letter on the list. There was no evidence of any closed-envelope technique to the alleged randomization. Moreover, in the protocol provided by Bezwoda, the first regimen listed was the conventional-dose CNV, presumably regimen A. Some of the records had an unsigned notation that the patient was randomized to arm A HDC, and these patients apparently did receive HDC, with the single exception of a patient who probably received nothing other than tamoxifen. However, one patient record had a notation signed by Seymour that the patient had been randomized to arm B to receive HDC. This statement would be consistent with the protocol in that the second regimen (presumably arm B) was HDC.

The 90-patient enrollment list provided by Bezwoda was written in the same handwriting throughout. Of the first 24 patients entered, only six were designated to have received HDC. Then, of the last 27 patients, only six were designated to have received the CNV regimen. It is unlikely that this sequence of treatment assignments could have occurred if the study were truly randomized.

Of the 90 patients allegedly enrolled onto this trial, some medical records were available for only 61. Fifty-seven of these patients were verified by a pathology report and/or clinical findings to have been diagnosed with breast cancer at some time. For the 29 other patients, not even a single document could be found that verified the patient had breast cancer or was treated at either of the two involved institutions. Eighteen of these names were designated to have received CNV and 11, HDC.

An attempt was made to determine eligibility of each patient using the criteria described in the published article.¹⁵ This determination was considerably hindered by the scarcity of information in the records, and definite eligibility could be judged only for a total of 27 patients (Table 1). Only nine (33%) of these 27 were verified as being eligible by all published¹⁵ criteria. Eighteen were established to be ineligible for reasons described in Table 1. An additional 14 patients were eligible by all criteria except that results of all blood tests and/or a multigated radionuclide cardiac scan were not available to establish eligibility fully. Twenty patients had insufficient information in the records, other than the diagnosis and age (but two did not even have a verifiable age), to establish eligibility.

Two patients receiving HDC had undergone the second cycle of HDC and had very low WBC and platelet counts when last seen shortly after this therapy. Given the available records, neither appeared to have had further follow-up, and they may have died soon thereafter. A third patient was registered on the log as receiving CNV, but she received one cycle of HDC instead. When last seen a week after this therapy, she was severely pancytopenic. There was no evidence she returned again, and she may also have died from toxicity.

Only seven of the 61 patients reviewed could be determined to be alive after 1995. One woman had received HDC in 1991 and was in continued complete remission (CR) when last seen in November 1997, and another had relapsed in 1995 but was still alive in October 1997. Two women had received HDC in 1992; one was alive and in CR in 1998, and the other apparently was still alive in 1999. Two others were alive in 1997 and 1998, but there were no records to establish they had ever received the therapy to which they were registered on the log. One patient was registered to HDC but did not receive it and was alive in April 2000.

Three record sets had a number of unsigned sequential chart entries in the same handwriting with facts at variance with information derived from other sources reviewed in the audit. Two of these patients are counted in Table 1 among the 25 who had received their assigned HDC.

Most of the 16 clinical trial articles published by Bezwoda et al that were assessed for a statement regarding IRB review contained the following statement verbatim or a minor variation: "All patients gave their informed consent and the study was performed in accordance with the principles of the Declaration of Helsinki and was approved by the Committee of Ethics and Human Research of the University of the Witwatersrand."³¹ For eight^31-38 of these articles there was no evidence that a protocol had ever been submitted to the IRB or that approval had been granted for the study. An additional study³⁹ contained only the generic phrase "the protocol was approved by the ethics committee at each of the participating centres." This statement was also untrue for this study³⁹ regarding participation at the University of the Witwatersrand.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
During the 1990s, the topic of HDC for treatment of various stages of breast cancer was surely the most controversial issue in the field of oncology. There were scientific debates in the medical literature, and 10 states even passed legislation⁴⁰ to guarantee such therapy would be covered by insurance carriers. Into this emotionally charged debate came this study conducted at a university center in Johannesburg. Although it involved only 90 patients, its influence was greatly magnified by the fact it was the first study to be reported in which there was a stated randomized comparison to a conventional-dose regimen. Many oncologists accepted the results as valid and believed that transplant therapy for metastatic breast cancer was now justified by the favorable results of a randomized study, the first of its kind published. The editorial²³ accompanying this article stated that "Bezwoda et al are to be commended for performing this study. The results are provocative and encouraging. However, caution must be exercised in interpreting these outcomes ... ."

The results of this audit show that the trial was not conducted in a scientifically acceptable manner. The protocol was apparently written 9 years after the study was started and only after another study by the same investigator was to be audited. No patient signed a consent form, and there is little evidence of true randomization of the patients. Records for 29 patients allegedly entered could not be found, despite a supreme effort to locate them. The records found for 20 patients had insufficient information—other than diagnosis and age—to establish eligibility for the study, and at least 18 other patients did not meet the stated eligibility criteria. Only three patients assigned to the CNV regimen could be verified to have received this regimen; some patients were treated with regimens and hormonal agents not consistent with the published¹⁵ information. There were at least three possible treatment-related deaths among patients receiving HDC, and survival of only seven of the 61 patients (some of whom may not have truly received the protocol therapy) could be verified beyond 1995. Not only could the published results not be validated by an audit of the patient records, but two statisticians identified some errors in the data and analysis presented in the article itself.^41,42

It is of interest that the first data derived from this trial were published in an ASCO abstract.¹⁶ This abstract indicates that "33 evaluable" patients had been entered as of the time of the abstract preparation (presumably late fall 1991), with 16 receiving HDC and 17 receiving CNV. According to the enrollment list provided by Bezwoda, of the first 33 patients (with the 33rd being entered on July 29, 1991), 23 patients were entered onto the CNV arm and only 10 were entered onto the HDC regimen, an unexplained inconsistency with the published abstract.¹⁶

The published article¹⁵ states that the metastatic disease study was conducted "in accordance with the principles of the Declaration of Helsinki" and that it was approved by the local IRB. The texts of eight other publications^31-38 also make these same statements, while one³⁹ just states that the study was approved by the IRB. Section 2 of the Declaration of Helsinki⁴³ states that "the design and performance of each experimental procedure involving human subjects should be clearly formulated in an experimental protocol which should be transmitted for consideration, comment and guidance to a specially appointed committee independent of the investigator . . . ." The audit verified that none of the 10 studies^15,31-39 assessed had been reviewed by the duly-constituted IRB, and Bezwoda himself admitted writing the protocol for the adjuvant study 9 years after the study was allegedly initiated.³⁰ Since the metastatic study protocol provided to the January audit team was identical in appearance to that of the adjuvant study, it likely was also written 9 years after the fact. Section 9 of the Declaration⁴³ states that the investigator "should then obtain the subject’s freely-given informed consent, preferably in writing." The auditors verified that no patient ever signed a consent form for the therapy allegedly administered under this study, and some may not even have been aware they were entered onto a study and had been randomized. Finally, Section 8 of the Declaration⁴³ states that "in publication of the results of his or her research, the physician is obliged to preserve the accuracy of the results." It is clear that none of these three sections of the Declaration of Helsinki were followed with specific reference to the metastatic study, and the statements in the multiple other assessed publications that the reported study did undergo review and approval by the "specially appointed" local committee are also untrue.

Bezwoda has published results from the metastatic study in nine abstracts or articles.^14-22 Table 2 delineates the drug doses published in these articles and compares them with the protocol provided. The inconsistency of the cyclophosphamide doses is noteworthy. Also noteworthy is the 1994 publication by Bezwoda et al.¹⁷ This article was received by the journal Hematological Oncology on February 24, 1994. Patient no. 90 (who is indicated to be the 45th patient to have received HDC) is said to have been entered onto the randomized trial exactly 1 year earlier (on February 25, 1993), according to the enrollment log. However, in the article¹⁷ submitted a year later, only 22 patients are described as receiving HDC as primary treatment for metastatic breast cancer at this institution, and one of them received a regimen that was different from the regimens listed in Table 2. It is of interest that the records search for the audit turned up an identical number of 22 patients (Table 1) who seemed to have received HDC.

This study had two coauthors on the first three public presentations of the data.^14-16 One of these coauthors (L. Seymour, MD) was present at the audit and was extraordinarily helpful to the audit team. It was apparent from the record review that she had been involved in the care of some of the patients, but there was no evidence she was involved in the analysis and reporting of data for this study.

We conclude that the multiple publications of this study of HDC in metastatic breast cancer do not report verifiable data. We have also ascertained that a total of nine other publications from this institution contain at least one untrue statement (ie, that approval by the University of the Witwatersrand’s Committee for Research on Human Subjects [the IRB] had been obtained for the study).

	APPENDIX

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
Illustrative Case Histories in the Patient Review

• According to the enrollment log, one patient was randomized to CNV in June 1991. However, no information could be found that she did receive such therapy, and in November 1991, she definitely received HDC.
  
• The birth date of a patient receiving HDC was recorded on the pathology report as being in January 1936, which would have made her age 55 at the time of registration in February 1991 and therefore ineligible. However, her age in the oncology ward records was written as 43 years.
  
• A patient registered onto the CNV regimen probably did receive this therapy, but the dates of therapy were widely disparate from the enrollment/randomization date on the log.
  
• A patient did indeed have metastatic breast cancer, but the last follow-up date in the records available was July 1990, 13 months before her alleged enrollment onto the study to receive CNV.
  
• A patient definitely received CNV, but the date of first therapy was approximately 9 months before the enrollment date on the log. Even then, she was ineligible for the study because she was approximately age 64 at the time of enrollment.
  
• A patient indicated on the enrollment log as receiving CNV was stated in the record as "ideally should be admitted for" HDC. She did not receive HDC, and during the same month she was supposed to be receiving CNV according to the enrollment log, she was started on single-agent therapy with epirubicin. There was no evidence she ever received CNV.
  
• A patient with locally advanced disease received cyclophosphamide, methotrexate, and fluorouracil (CMF)initially, followed by a mastectomy and adjuvant radiation during the same month she was allegedly randomized to CNV. Postmastectomy chemotherapy was also given, and she received a total of 12 cycles of CMF. There was no evidence she ever received CNV.
  
• One patient treated by Dr Seymour (and specifically remembered by her) was "randomized" to CNV but declined this therapy. She instead chose HDC and did undergo this therapy as documented by her records that were audited. Her name was not on the enrollment list.
  
• One patient designated as receiving CNV received three cycles of this therapy and then three cycles of the CAF regimen.
  
• A patient treated with HDC received the first cycle (with the treatment orders written and signed by the principal investigator) in doses of cyclophosphamide 50 mg/kg, mitoxantrone 40 mg/m², and etoposide 1.5 g/m², which were not the published drug doses. The second cycle was given in the doses specified in the 1995 publication.¹⁵
  
• One patient registered as receiving CNV was begun on cyclophosphamide, epirubicin, and fluorouracil (CEF) the same month she was supposed to have been enrolled onto the study. She did not receive CNV until 16 months later, after progressing further despite some intervening hormonal therapies.
  
• A patient registered onto the CNV regimen was only seen initially in the "chemo clinic" 4 months after the randomization date and at that time received CAF. Another patient registered as receiving CNV also received only CAF beginning approximately the same month as the enrollment date.
  
• One patient had been permanently institutionalized for mental retardation. One note in her records indicated that she was not a suitable candidate for HDC or systemic chemotherapy and should only receive tamoxifen as therapy for her cancer. However, she was recorded on the enrollment list as receiving HDC, and there were notations in the records indicating she did receive two cycles of such therapy. The auditors believe this latter information was not valid.
  
• One patient registered to receive CNV underwent HDC with one cycle of a combination regimen of carmustine, carboplatin, and cyclophosphamide 1 week after the alleged CNV enrollment.
  
• One patient was treated with a single cycle of HDC in April 1992 using the protocol regimen and had a "good response." However, she was recorded on the enrollment log as being randomized on July 31, 1992. Five days after this date, she was noted in the medical records to have progressed since the prior transplant. She was then treated with HDC once again on August 6, 1992, but this time she was given a combination of cyclophosphamide, carmustine, and carboplatin. She was discharged on September 25, 1992, and died 25 days later outside the hospital, possibly due to transplant-related effects.
  
• One patient was recorded on the log as being enrolled in August 1992 and randomized to HDC. The only information available for this patient was the fact she was diagnosed with breast cancer in 1988 and was alive to receive a prescription in May 1998. There were no records of treatment available, and an attempt at telephone contact during the audit failed.
  
• A patientwas diagnosed as having breast cancer in 1989 and received adjuvant chemotherapy. She was recorded on the log as being randomized to HDC in September 1992. However, there was no information in the available records regarding her treatment. She was last known to be alive in October 1997.
  
• A patient was treated for breast cancer in 1984 and 1985. The last available information in the record was dated November 1985. She was recorded on the log as being randomized to CNV in October 1992. There was no information she was even alive after 1985.
  
• One patient was registered as receiving CNV. She was ineligible because of her age (56 years) and the fact that she had a significant anemia. There were no other records available to document therapy.
  
• One patient was registered to HDC on the log in December 1992. At that time, she was likely ineligible for the protocol because she had not been off adjuvant chemotherapy for at least 6 months. The only record of transplant therapy was information that in November 1993 she was treated with carmustine, ifosfamide, etoposide, and carboplatin in a high-dose regimen.
  
• One patient did indeed undergo HDC 1 week after the registration date on the log and assignment to HDC, but the drug doses were lower (ie, cyclophosphamide 2.2 g/m², mitoxantrone 30 mg/m², and etoposide 1.8 g/m²) than those in the published regimen.¹⁵ She declined further therapy after one cycle.
  
• One patient was registered as receiving HDC in January 1993. However, she was ineligible because of age (54 years), and exactly 30 days after the registration date, she was begun on treatment with epirubicin and vinblastine, with the drug orders being signed by the principal investigator. She received five cycles of this regimen and was alive in April 2000 according to the medical records.
  
• One patient was registered in February 1993 as receiving HDC, but all the records indicated she received CNV beginning in April 1993.
  
• One patient registered to CNV in February 1993 did indeed receive the CNV regimen for a total of eight cycles, but she did not begin this therapy until April 1993.
  

	ACKNOWLEDGMENTS

 
The authors thank Peter Cleaton-Jones, MB BCh, PhD, DSc(Dent), DPH, Assistant Dean (Research), Faculty of Health Sciences of the University of the Witwatersrand, for his great assistance in the process of this audit. Salome Liebenberg of the department of radiation oncology was also of great help in locating patient records.

	NOTES

 
R.B.W. and G.G.G. are independent consultants and provide research auditing services for various entities.

Published online before print at http://www.asco.org/jco/ascoindex.htm.

An electronic version of this article, with brief clinical details of selected patients in an Appendix, is available at the following Internet site: http://www.jco.org.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS AND SCOPE OF... RESULTS DISCUSSION APPENDIX REFERENCES

 
1. Peters WP, Shpall EJ, Jones RB, et al: High-dose combination alkylating agents with bone marrow support as initial treatment for metastatic breast cancer. J Clin Oncol 6: 1368-1376, 1988[Abstract/Free Full Text]

2. Peters WP, Ross M, Vredenburgh JJ, et al: High-dose chemotherapy and autologous bone marrow support as consolidation after standard-dose adjuvant therapy for high-risk primary breast cancer. J Clin Oncol 11: 1132-1143, 1993[Abstract/Free Full Text]

3. Antman KH, Rowlings PA, Vaughan WP, et al: High-dose chemotherapy with autologous hematopoietic stem-cell support for breast cancer in North America. J Clin Oncol 15: 1870-1879, 1997[Abstract/Free Full Text]

4. Gianni AM, Siena S, Bregni M, et al: Efficacy, toxicity, and applicability of high-dose sequential chemotherapy as adjuvant treatment in operable breast cancer with 10 or more involved axillary nodes: Five-year results. J Clin Oncol 15: 2312-2321, 1997[Abstract/Free Full Text]

5. Smigel K: Women flock to ABMT for breast cancer without final proof. J Natl Cancer Inst 87: 952-955, 1995[Free Full Text]

6. Vahdat L, Antman K: High-dose therapy for breast cancer. Blood Rev 9: 191-200, 1995[Medline]

7. Smith GA, Henderson IC: High-dose chemotherapy (HDC) with autologous bone marrow transplantation (ABMT) for the treatment of breast cancer: The jury is still out, in DeVita VT, Hellman S, Rosenberg SA (eds): Important Advances in Oncology 1995. Philadelphia PA: J.B. Lippincott Co, 1995, pp 201-214

8. Siu LL, Tannock IF: Chemotherapy dose escalation: Case unproven. J Clin Oncol 15: 2765-2768, 1997[Medline]

9. Weiss RB: The randomized trials of dose-intensive therapy for breast cancer: What do they mean for patient care and where do we go from here? Oncologist 4: 450-458, 1999[Abstract/Free Full Text]

10. Canellos GP: Selection bias in trials of transplantation for metastatic breast cancer: Have we picked the apple before it was ripe? J Clin Oncol 15: 3169-3170, 1997[Medline]

11. Glück S, Stewart D: High-dose chemotherapy in breast cancer: Out of favor but not out of promise. Bone Marrow Transplant 25: 1017-1019, 2000[Medline]

12. Antman KH: High-dose chemotherapy in breast cancer: The end of the beginning? Biol Blood Marrow Transplant 6: 469-475, 2000[Medline]

13. Rahman ZU, Frye DK, Buzdar AU, et al: Impact of selection process on response rate and long-term survival of potential high-dose chemotherapy candidates treated with standard-dose doxorubicin-containing chemotherapy in patients with metastatic breast cancer. J Clin Oncol 15: 3171-3177, 1997[Abstract]

14. Bezwoda WR, Seymour L, Dansey RD: High dose chemotherapy with hematopoetic rescue as primary treatment for metastatic breast cancer: A randomised trial. Eur J Cancer 31A: S76, 1995 (suppl 5, abstr 349)

15. Bezwoda WR, Seymour L, Dansey RD: High-dose chemotherapy with hematopoietic rescue as primary treatment for metastatic breast cancer: A randomized trial. J Clin Oncol 13: 2483-2489, 1995[Abstract]

16. Bezwoda WR, Seymour L, Vorobiof DA: High dose cyclophosphamide, mitoxantrone, and VP-16 (HD-CNV) as first-line treatment for metastatic breast cancer. Proc Am Soc Clin Oncol 11: 64, 1992 (abstr 81)

17. Bezwoda WR, Dansey R, Seymour L, et al: Non-cryopreserved, limited number (1 or 2) peripheral blood progenitor cell (PBPC) collections following GCSF administration provide adequate hematologic support for high dose chemotherapy. Hematol Oncol 12: 101-110, 1994[Medline]

18. Bezwoda WR: In vivo purging of peripheral stem cell products by high-dose CNVp: Results from a randomised study comparing conventional dose chemotherapy (CNV) with double high-dose CNVp chemotherapy. Proc Am Soc Clin Oncol 16: 115a, 1997 (abstr 404)

19. Bezwoda WR: High-dose chemotherapy with hematopoietic rescue in breast cancer: From theory to practice. Cancer Chemother Pharmacol 40: S79-S87, 1997 (suppl)

20. Bezwoda WR: High-dose chemotherapy with haematopoietic rescue: Application to primary management of metastatic breast cancer. Eur J Cancer Care 6: 10-15, 1997 (suppl 1)

21. Bezwoda WR: Primary high dose chemotherapy for metastatic breast cancer: Update and analysis of prognostic factors. Proc Am Soc Clin Oncol 17: 115a, 1998 (abstr 445)

22. Bezwoda WR: High dose chemotherapy with haematopoietic rescue in breast cancer. Hematol Cell Ther 41: 58-65, 1999[Medline]

23. Kennedy MJ: High-dose chemotherapy of breast cancer: Is the question answered? J Clin Oncol 13: 2477-2479, 1995 (editorial)[Medline]

24. Weiss RB, Rifkin RM, Stewart FM, et al: High-dose chemotherapy for high-risk primary breast cancer: An on-site review of the Bezwoda study. Lancet 355: 999-1003, 2000[Medline]

25. Canellos GP: The policing of clinical trials. J Clin Oncol 18: 2353, 2000 (editorial)[Free Full Text]

26. Weiss RB, Vogelzang NJ, Peterson BA, et al: A successful system of scientific data audits for clinical trials: A report from the Cancer and Leukemia Group B. JAMA 270: 459-464, 1993[Abstract]

27. Weiss RB: Systems of protocol review, quality assurance, and data audit. Cancer Chemother Pharmacol 42: S88-S92, 1998 (suppl)

28. Office of the Deputy Registrar (Research): Information on Research Grants. Johannesburg South Africa: University of the Witwatersrand, 1987

29. Ethical Considerations in Medical Research, Revised Edition. Parow South Africa: South African Medical Research Council, 1987

30. Cleaton-Jones P: Policing research: Remember the ethics. MRC News 31:6-7, October 2000

31. Bezwoda WR, Seymour L, Ariad S: High-dose etoposide in treatment of metastatic breast cancer. Oncology 49: 104-107, 1992[Medline]

32. Bezwoda WR, Weaving A, Kew M, et al: Combination chemotherapy of hepatocellular cancer: Comparison of Adriamycin + VM 26 + 5-fluorouracil with mAMSA + VM 26 + 5-fluorouracil. Oncology 44: 207-209, 1987[Medline]

33. Bezwoda WR, Bernasconi C, Hutchinson RM, et al: Mito-xantrone for refractory and relapsed acute leukemia. Cancer 66: 418-422, 1990[Medline]

34. Bezwoda WR: Treatment of stage D2 prostatic cancer refractory to or relapsed following castration plus oestrogens: Comparison of aminoglutethimide plus hydrocortisone with mexdroxyprogesterone acetate plus hydrocortisone. Br J Urol 66: 196-201, 1990[Medline]

35. Ariad S, Seymour L, Bezwoda WR: Platelet-derived growth factor (PDGF) in plasma of breast cancer patients: Correlation with stage and rate of progression. Breast Cancer Res Treat 20: 11-17, 1991[Medline]

36. Bezwoda WR, Bezwoda MA, Seymour L, et al: Nope for relapsed aggressive diffuse non-Hodgkin’s lymphoma. Leuk Lymphoma 10: 329-333, 1993[Medline]

37. Seymour L, Meyer K, Bezwoda WR: Hormone priming in breast cancer: Oestrogen priming has a detrimental effect on response in oestrogen receptor-negative patients. Eur J Cancer 29A: 1495-1496, 1993

38. Vorobiof DA, Bezwoda WR: A randomised trial of vindesine plus interferon-alpha2b compared with interferon-alpha2b or vindesine alone in the treatment of advanced malignant melanoma. Eur J Cancer 30A: 797-800, 1994

39. Brouwer RE, Vellenga E, Zwinderman KH, et al: Phase III efficacy study of interleukin-3 after autologous bone marrow transplantation in patients with malignant lymphoma. Br J Haematol 106: 730-736, 1999[Medline]

40. Division of Cancer Prevention and Control: State Laws Relating to Breast Cancer: Legislative Summary January 1949 to May 2000. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, http://www.cdc.gov/cancer/nbccedp/bccpdfs/BCLaws.pdf

41. Vickers A, Christos P: Bezwoda: Evidence of fabrication in original article. J Clin Oncol 18: 2933, 2000 (letter)[Free Full Text]

42. Vickers A, Christos P: Re: "Bezwoda: Evidence of fabrication in original article." J Clin Oncol 18: 3875, 2000 (letter)[Free Full Text]

43. Declaration of Helsinki: Recommendations guiding medical doctors in biomedical research involving human subjects. Med J Aust 1:206-207, 1976

Submitted January 2, 2001; accepted March 5, 2001.

This article has been cited by other articles:

				J. P. Ioannidis and T. A Trikalinos An exploratory test for an excess of significant findings Clinical Trials, June 1, 2007; 4(3): 245 - 253. [Abstract] [PDF]

				P. Schmid, W. Schippinger, T. Nitsch, G. Huebner, V. Heilmann, W. Schultze, H. Hausmaninger, M. Wischnewsky, and K. Possinger Up-Front Tandem High-Dose Chemotherapy Compared With Standard Chemotherapy With Doxorubicin and Paclitaxel in Metastatic Breast Cancer: Results of a Randomized Trial J. Clin. Oncol., January 20, 2005; 23(3): 432 - 440. [Abstract] [Full Text] [PDF]

				R. C. F. Leonard, M. Lind, C. Twelves, R. Coleman, S. van Belle, C. Wilson, J. Ledermann, I. Kennedy, P. Barrett-Lee, T. Perren, et al. Conventional Adjuvant Chemotherapy Versus Single-Cycle, Autograft-Supported, High-Dose, Late-Intensification Chemotherapy in High-Risk Breast Cancer Patients: A Randomized Trial J Natl Cancer Inst, July 21, 2004; 96(14): 1076 - 1083. [Abstract] [Full Text] [PDF]

				L. Gianni High-dose chemotherapy for breast cancer: any use for it? Ann. Onc., May 1, 2002; 13(5): 650 - 652. [Full Text] [PDF]

				J. R. Gralow and R. B. Livingston University of Washington High-Dose Cyclophosphamide, Mitoxantrone, and Etoposide Experience in Metastatic Breast Cancer: Unexpected Cardiac Toxicity J. Clin. Oncol., September 15, 2001; 19(18): 3903 - 3904. [Full Text] [PDF]

				R. P. Abratt, D. A. Vorobiof, L. Goedhals, B. Rappaport, P. Ruff, and C. F. Slabber Scientific Fraud and International Research J. Clin. Oncol., August 1, 2001; 19(15): 3592 - 3592. [Full Text]

				L. Norton High-Dose Chemotherapy for Breast Cancer: ""How Do You Know?"" J. Clin. Oncol., June 1, 2001; 19(11): 2769 - 2770. [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 Weiss, R. B. Articles by Hudis, C. A. Search for Related Content PubMed PubMed Citation Articles by Weiss, R. B. Articles by Hudis, C. A.