Through the Avastin Looking Glass
06/11/06 08:28 AM
Published to Retinadialogue.com in summer ‘06
The following has been called “great” (Phil Rosenfeld), “very important”(Gary Abrams), “right on the mark”(Sandy Brucker), and “the right thing to do” (Anne Fung) by several of the leaders in our specialty. It has been criticized as being “biased” by some who have financial relationships with one or more corporations. It has been submitted to several retinal publications, but rejected for being “too controversial” for their advertisers (I wonder who?). It is controversial, but it reflects concerns raised many times and in many forums in the retinal world. I have no desire to be a pariah or to disrespect the many fine doctors and researchers in the retinal field. Still, I do believe that these and future questions need to be raised and discussed. This is therefore an attempt to clarify rumors and innuendos that have been circulated about Avastin and to promote discussion about how we control or cede control of our practice of retinal medicine. I have no relationship of any kind with any pharmaceutical company.
Through the Avastin looking glass
I think I may have slipped through the looking glass and down the rabbit hole into a surreal world.
It seems that every time I turn around, I see an interview or a meeting or a flyer about how great PDT or Macugen is because patients lose fewer than 15 letters of vision. The participants seem to assume that Avastin doesn’t exist or that AMD treatment needs a less effective but more expensive combination drug for “maintenance” when those same patients could have actually gained vision safely with Avastin monotherapy at a fraction of the cost. Then it struck me. Many of these speakers and interviewees are paid either by Novartis or Pfizer, both of whom might wish that Avastin didn’t exist. Brief reference to financial relationships can be found in the fine print, but the façade is always that of an objective, scientific discussion impartially considering all of the facts. It is really no better in the equally surreal evolution of the Avastin/Lucentis controversy. Based on Genentech’s apparent actions to increase price and limit availability of Avastin and the publicity for Lucentis which also seems to assume that Avastin never existed, one could speculate that even they would like ophthalmic use of Avastin to disappear as they promote its analog at an exponentially higher price.
It is the potential for marketing plans to guide the direction of medical care and our responsibility to maintain independence and objectivity as we provide care for our patients that concerns me. This is where we have begun to stray and where we need to refocus and guide corporate sponsorship of care rather than have that care guided by corporate interests. The Avastin story provides an opportunity to explore trends that have been growing for some time. I have no desire to disparage or disrespect any individual or corporation but rather to simply raise points of concern that have entered the world-wide retinal dialogue.
A good example of misinformation came across all of our desks this past summer: A glossy Pfizer flyer talking about thromboembolism in the elderly, written as if it is a foregone conclusion that nonselective VEGF inhibition causes stroke. Pfizer has based its recent promotional campaign on implicating Avastin (and Lucentis) for incidental, probably unrelated vascular disease common in the elderly patient population. As evidence for a causal relationship, this flyer referred to van Wijngaarden et al’s 1 observation that if your instruments are sensitive enough, you might detect traces of systemic Avastin after a 1.25 mg intraocular injection. This is supposed to be the smoking gun. Are we to ignore the fact that oncologists infuse 500 mg or more of Avastin directly into the blood stream every 2 weeks with virtual impunity? They consider the therapy “very effective and very benign” (except for possible interference with wound healing). Of course, the cancer patients may have little to lose since their options are limited, but contrary to the implication, the now sizable and ever-growing ophthalmic literature 2-12 on the safety and efficacy of intravitreal Avastin yields no provable adverse effect. The even more voluminous Avastin safety survey13 also supports a high margin of safety.
As I understand it, the stroke story is based on advisory letters that Genentech released in August 2004 and January 2005. Based on a meta-analysis of cancer patients, they warned that in patients with disseminated colorectal cancer (and only in that prothrombotic patient population), the combination of high dose IV Avastin and high dose IV 5-Fluorouracil may increase the annual rate of all thrombotic events from a high 1.9 percent to a slightly higher 4.4 percent. At the time, Dr. Hal Barron, Chief Medical Officer for Genentech was quoted in Nature Biotechnology 14 saying, “It isn’t absolutely known whether the effects seen with Avastin can be generalized beyond the population of metastatic cancer patients receiving 5-FU….The implications of it being unique to 5-FU or to colon cancer would be very important for clinicians to understand.” The thromboembolism advisory was directed only to patients with disseminated colorectal cancer and only with the act of combining IV 5-FU with IV Avastin. The message is now different. Genentech was cited on CNBC and elsewhere in December 2005 as indicating that physicians and patients should just wait for Lucentis because using Avastin in the eye is conclusively “unsafe”. Could financial factors have had any role in this change of heart?
I am aware of no other evidence implicating Avastin in thromboembolism. I have been told by every oncology colleague I have asked that drawing an analogy from this single advisory and ignoring the volumes of safety data from the rest of the oncology literature is unreasonable, especially considering the differences:
It is not surprising that the drug companies are fighting acceptance of Avastin for ocular use. They have worked hard to establish a status quo where we are all supposed to accept, even embrace, the concept of medicines in the growing field of retinal chemotherapy costing $1,000 to $1,500 (or even $18,250) per dose. We are furthermore supposed to consider 3 lines of additional vision loss a “successful response”. Unfortunately for this status quo, Avastin and its analog Lucentis have redefined “successful response” as one where most patients gain vision quickly and unequivocally, especially if we exclude patients who have already lost vision during Macugen or PDT therapy.
Then there is the question of cost-effectiveness. The cost differences between therapies are stark. According to published protocols, the baseline cost per case (just for the drug and not including triamcinolone combinations) is roughly:
As the saying goes, “you do the math”. Of course, there are variables in comparing Lucentis and Avastin. Avastin sometimes results in stability after a single dose and other times may require multiple injections. Similarly, it might not be necessary to give Lucentis every month for long durations, although the PIER data demonstrates that decreasing to quarterly injections results in resumed progressive loss of vision. At least in the short term, Lucentis seems to have a beneficial effect similar to Avastin, but at potentially 133 times the cost per case, it would need to be greatly superior to be justifiable. Indeed, the question of whether there is any role for Lucentis at all can only be answered by a head to head comparison with Avastin. If this is not done prospectively, physicians will have the appropriate responsibility to make recommendations based on the data that is available. There is enough experience with both to be confident that they are similarly safe. The simple fact is that with volumes of data published by reputable sources, Avastin sets the de facto standard for the combination of safety, efficacy, and cost against which all future AMD treatments including Lucentis must be compared. Unless Lucentis is priced according to this standard, its release will become irrelevant to patient care and a step backwards in cost-effectiveness. Doubling the cost of Avastin and/or making it less accessible are hardly appropriate ways to make Lucentis seem more acceptable by comparison.
None of us want our patients harmed, but there is no current proof of adverse effects caused by intravitreal Avastin as compared to Lucentis or any other therapy. We must remember that losing vision from an inferior treatment and/or adding unnecessary additional cost is harm too. We don’t know that Avastin (or Lucentis) never causes adverse effects, but the empirical data supporting the safety of both forms of this molecule are very compelling and reassuring. We certainly shouldn’t ignore risk, but when those with a financial interest are willing to manufacture or distort data to wage a campaign of fear, we owe it to our patients to stand up and do our best to act on truth rather than “urban legend”.
It is unfortunate that since the introduction of PDT, some of our colleagues have literally sold their support to corporations to promote certain treatments to the exclusion of newer and better alternatives. Whether it is stock options, money, funding for chaired professorships, or just travel expenses to exotic destinations, such remuneration can adversely affect credibility and perhaps objectivity as well. Paid endorsement, in any form, needs to be clearly and completely represented as such. There are many physicians of impeccable integrity who work with various corporations, and they adequately disclose those relationships. That is not the problem. The problem lies with those willing to sell opinions and whose relationships are more nebulously described. Research bias in company-funded studies, even if not intended, is another concern. This is not unique to ophthalmology. Evidence of favorable bias in company-funded research was recently reported in the British Medical Journal.15 Of course, we deeply appreciate the efforts of pharmaceutical companies to develop new drugs. We appropriately want to help with research and advice, and it is appropriate that such help be compensated. Compensation for services rendered is perfectly acceptable; however, the moment a physician assumes a vested interest in the success of a company or product (such as accepting stock options) or accepts compensation or reimbursement specifically to endorse a product or disparage a competitor, that physician can then act only as an agent for his or her “employer”. They must recuse themselves from independent discussion. Many of us have been offered trips or stock options too, but most of us have declined when a vested or promotional interest is involved. We all make pretty good livings without it, and we need to be able to say what we believe to be true without having our motives questioned. The case of a physician being a principle in a company is of course above board, easily understood, and nonambiguous.
According to all available data, Avastin appears to be extremely safe and effective in the majority of patients. Judging from the Spring 2006 ASRS Avastin survey, this appears to be the strong consensus of the retina community. The retinadialogue.com surveys show over 80% of respondents continuing to use Avastin 2 months after the release of Lucentis, supporting the consensus that Avastin is a legitimate standard of care. We need to decide what to do; with Avastin of course but also with the broader questions that are raised by the Avastin issue. We should value the great work Genentech has done in developing Avastin; however, we should be mystified when, in the face of compelling data from reliable sources, the company’s position continues to be that ‘it doesn’t work’ (Miami 2006 Antiangiogenesis Meeting). We should wonder about the motivations for developing another version of the same drug. Was it to improve it, or was it to circumvent the “pricing problem”? We should consider our own responsibility in choosing between two forms of the same drug with similar safety and efficacy but radically different cost. We should be circumspect about taking corporate stock or money, especially if it leads us to do things like redefine “success” to include three lines of failure in order to make marginal treatments look acceptable. We should be alarmed when corporations wage a disinformation campaign based on data that a first year medical student could debunk. We should stand up for our patients and regain our credibility with them, with payers, with the government, and indeed with eachother by working with facts instead of spinning them. We should start acting like doctors again instead of corporate agents.
We need to decide.
References:
The following has been called “great” (Phil Rosenfeld), “very important”(Gary Abrams), “right on the mark”(Sandy Brucker), and “the right thing to do” (Anne Fung) by several of the leaders in our specialty. It has been criticized as being “biased” by some who have financial relationships with one or more corporations. It has been submitted to several retinal publications, but rejected for being “too controversial” for their advertisers (I wonder who?). It is controversial, but it reflects concerns raised many times and in many forums in the retinal world. I have no desire to be a pariah or to disrespect the many fine doctors and researchers in the retinal field. Still, I do believe that these and future questions need to be raised and discussed. This is therefore an attempt to clarify rumors and innuendos that have been circulated about Avastin and to promote discussion about how we control or cede control of our practice of retinal medicine. I have no relationship of any kind with any pharmaceutical company.
Through the Avastin looking glass
I think I may have slipped through the looking glass and down the rabbit hole into a surreal world.
It seems that every time I turn around, I see an interview or a meeting or a flyer about how great PDT or Macugen is because patients lose fewer than 15 letters of vision. The participants seem to assume that Avastin doesn’t exist or that AMD treatment needs a less effective but more expensive combination drug for “maintenance” when those same patients could have actually gained vision safely with Avastin monotherapy at a fraction of the cost. Then it struck me. Many of these speakers and interviewees are paid either by Novartis or Pfizer, both of whom might wish that Avastin didn’t exist. Brief reference to financial relationships can be found in the fine print, but the façade is always that of an objective, scientific discussion impartially considering all of the facts. It is really no better in the equally surreal evolution of the Avastin/Lucentis controversy. Based on Genentech’s apparent actions to increase price and limit availability of Avastin and the publicity for Lucentis which also seems to assume that Avastin never existed, one could speculate that even they would like ophthalmic use of Avastin to disappear as they promote its analog at an exponentially higher price.
It is the potential for marketing plans to guide the direction of medical care and our responsibility to maintain independence and objectivity as we provide care for our patients that concerns me. This is where we have begun to stray and where we need to refocus and guide corporate sponsorship of care rather than have that care guided by corporate interests. The Avastin story provides an opportunity to explore trends that have been growing for some time. I have no desire to disparage or disrespect any individual or corporation but rather to simply raise points of concern that have entered the world-wide retinal dialogue.
A good example of misinformation came across all of our desks this past summer: A glossy Pfizer flyer talking about thromboembolism in the elderly, written as if it is a foregone conclusion that nonselective VEGF inhibition causes stroke. Pfizer has based its recent promotional campaign on implicating Avastin (and Lucentis) for incidental, probably unrelated vascular disease common in the elderly patient population. As evidence for a causal relationship, this flyer referred to van Wijngaarden et al’s 1 observation that if your instruments are sensitive enough, you might detect traces of systemic Avastin after a 1.25 mg intraocular injection. This is supposed to be the smoking gun. Are we to ignore the fact that oncologists infuse 500 mg or more of Avastin directly into the blood stream every 2 weeks with virtual impunity? They consider the therapy “very effective and very benign” (except for possible interference with wound healing). Of course, the cancer patients may have little to lose since their options are limited, but contrary to the implication, the now sizable and ever-growing ophthalmic literature 2-12 on the safety and efficacy of intravitreal Avastin yields no provable adverse effect. The even more voluminous Avastin safety survey13 also supports a high margin of safety.
As I understand it, the stroke story is based on advisory letters that Genentech released in August 2004 and January 2005. Based on a meta-analysis of cancer patients, they warned that in patients with disseminated colorectal cancer (and only in that prothrombotic patient population), the combination of high dose IV Avastin and high dose IV 5-Fluorouracil may increase the annual rate of all thrombotic events from a high 1.9 percent to a slightly higher 4.4 percent. At the time, Dr. Hal Barron, Chief Medical Officer for Genentech was quoted in Nature Biotechnology 14 saying, “It isn’t absolutely known whether the effects seen with Avastin can be generalized beyond the population of metastatic cancer patients receiving 5-FU….The implications of it being unique to 5-FU or to colon cancer would be very important for clinicians to understand.” The thromboembolism advisory was directed only to patients with disseminated colorectal cancer and only with the act of combining IV 5-FU with IV Avastin. The message is now different. Genentech was cited on CNBC and elsewhere in December 2005 as indicating that physicians and patients should just wait for Lucentis because using Avastin in the eye is conclusively “unsafe”. Could financial factors have had any role in this change of heart?
I am aware of no other evidence implicating Avastin in thromboembolism. I have been told by every oncology colleague I have asked that drawing an analogy from this single advisory and ignoring the volumes of safety data from the rest of the oncology literature is unreasonable, especially considering the differences:
- ♦ We are using 1.25 mg rather than 500 to 1000 mg or more
- ♦ We are injecting into the sequestered space of the eye rather than directly into the systemic circulation
- ♦ We are treating every 2 months or so rather than every 2 weeks
- ♦ We are not using 5-FU
- ♦ And of course, generally speaking our patients don’t have disseminated colorectal cancer. Ironically, there are cases of patients’ wet AMD improving during systemic Avastin therapy for cancer!
It is not surprising that the drug companies are fighting acceptance of Avastin for ocular use. They have worked hard to establish a status quo where we are all supposed to accept, even embrace, the concept of medicines in the growing field of retinal chemotherapy costing $1,000 to $1,500 (or even $18,250) per dose. We are furthermore supposed to consider 3 lines of additional vision loss a “successful response”. Unfortunately for this status quo, Avastin and its analog Lucentis have redefined “successful response” as one where most patients gain vision quickly and unequivocally, especially if we exclude patients who have already lost vision during Macugen or PDT therapy.
Then there is the question of cost-effectiveness. The cost differences between therapies are stark. According to published protocols, the baseline cost per case (just for the drug and not including triamcinolone combinations) is roughly:
- ♦ PDT- $1500.00 per dose every 3 months for 2 to 3 years = $12,000 to 18,000
- ♦ Macugen- $1000.00 per dose every 6 weeks for 2 years = $17,000
- ♦ Lucentis- $1950.00 per dose monthly for 2 years = $48,000!
- ♦ Avastin- ~$30.00 per dose every two months for two years (a frequency and duration that may be more than is necessary) = $360
As the saying goes, “you do the math”. Of course, there are variables in comparing Lucentis and Avastin. Avastin sometimes results in stability after a single dose and other times may require multiple injections. Similarly, it might not be necessary to give Lucentis every month for long durations, although the PIER data demonstrates that decreasing to quarterly injections results in resumed progressive loss of vision. At least in the short term, Lucentis seems to have a beneficial effect similar to Avastin, but at potentially 133 times the cost per case, it would need to be greatly superior to be justifiable. Indeed, the question of whether there is any role for Lucentis at all can only be answered by a head to head comparison with Avastin. If this is not done prospectively, physicians will have the appropriate responsibility to make recommendations based on the data that is available. There is enough experience with both to be confident that they are similarly safe. The simple fact is that with volumes of data published by reputable sources, Avastin sets the de facto standard for the combination of safety, efficacy, and cost against which all future AMD treatments including Lucentis must be compared. Unless Lucentis is priced according to this standard, its release will become irrelevant to patient care and a step backwards in cost-effectiveness. Doubling the cost of Avastin and/or making it less accessible are hardly appropriate ways to make Lucentis seem more acceptable by comparison.
None of us want our patients harmed, but there is no current proof of adverse effects caused by intravitreal Avastin as compared to Lucentis or any other therapy. We must remember that losing vision from an inferior treatment and/or adding unnecessary additional cost is harm too. We don’t know that Avastin (or Lucentis) never causes adverse effects, but the empirical data supporting the safety of both forms of this molecule are very compelling and reassuring. We certainly shouldn’t ignore risk, but when those with a financial interest are willing to manufacture or distort data to wage a campaign of fear, we owe it to our patients to stand up and do our best to act on truth rather than “urban legend”.
It is unfortunate that since the introduction of PDT, some of our colleagues have literally sold their support to corporations to promote certain treatments to the exclusion of newer and better alternatives. Whether it is stock options, money, funding for chaired professorships, or just travel expenses to exotic destinations, such remuneration can adversely affect credibility and perhaps objectivity as well. Paid endorsement, in any form, needs to be clearly and completely represented as such. There are many physicians of impeccable integrity who work with various corporations, and they adequately disclose those relationships. That is not the problem. The problem lies with those willing to sell opinions and whose relationships are more nebulously described. Research bias in company-funded studies, even if not intended, is another concern. This is not unique to ophthalmology. Evidence of favorable bias in company-funded research was recently reported in the British Medical Journal.15 Of course, we deeply appreciate the efforts of pharmaceutical companies to develop new drugs. We appropriately want to help with research and advice, and it is appropriate that such help be compensated. Compensation for services rendered is perfectly acceptable; however, the moment a physician assumes a vested interest in the success of a company or product (such as accepting stock options) or accepts compensation or reimbursement specifically to endorse a product or disparage a competitor, that physician can then act only as an agent for his or her “employer”. They must recuse themselves from independent discussion. Many of us have been offered trips or stock options too, but most of us have declined when a vested or promotional interest is involved. We all make pretty good livings without it, and we need to be able to say what we believe to be true without having our motives questioned. The case of a physician being a principle in a company is of course above board, easily understood, and nonambiguous.
According to all available data, Avastin appears to be extremely safe and effective in the majority of patients. Judging from the Spring 2006 ASRS Avastin survey, this appears to be the strong consensus of the retina community. The retinadialogue.com surveys show over 80% of respondents continuing to use Avastin 2 months after the release of Lucentis, supporting the consensus that Avastin is a legitimate standard of care. We need to decide what to do; with Avastin of course but also with the broader questions that are raised by the Avastin issue. We should value the great work Genentech has done in developing Avastin; however, we should be mystified when, in the face of compelling data from reliable sources, the company’s position continues to be that ‘it doesn’t work’ (Miami 2006 Antiangiogenesis Meeting). We should wonder about the motivations for developing another version of the same drug. Was it to improve it, or was it to circumvent the “pricing problem”? We should consider our own responsibility in choosing between two forms of the same drug with similar safety and efficacy but radically different cost. We should be circumspect about taking corporate stock or money, especially if it leads us to do things like redefine “success” to include three lines of failure in order to make marginal treatments look acceptable. We should be alarmed when corporations wage a disinformation campaign based on data that a first year medical student could debunk. We should stand up for our patients and regain our credibility with them, with payers, with the government, and indeed with eachother by working with facts instead of spinning them. We should start acting like doctors again instead of corporate agents.
We need to decide.
References:
- 1. Inhibitors of ocular neovascularization: promises and potential problems.: van Wijngaarden P, Coster DJ, Williams KA, JAMA 2005; 293:1509-13
- 2. THE ROLE OF VASCULAR ENDOTHELIAL GROWTH FACTOR IN OCULAR HEALTH AND DISEASE, ANTHONY P. ADAMIS, MD, DAVID T. SHIMA, PHD RETINA 25:111–118, 2005
- 3. INTRAVITREAL BEVACIZUMAB (AVASTIN) TREATMENT OF PROLIFERATIVE DIABETIC RETINOPATHY COMPLICATED BY VITREOUS HEMORRHAGE; RICHARD F. SPAIDE, MD, YALE L. FISHER, MD RETINA 26:275–278, 2006
- 4. TESTING INTRAVITREAL TOXICITY OF BEVACIZUMAB (AVASTIN); ROBERTA P. A. MANZANO, MD, GHOLAM A. PEYMAN, MD, PALWASHA KHAN, MD, MUHAMET KIVILCIM, MD, RETINA 26:257–261, 2006
- 5. REGRESSION OF RETINAL AND IRIS NEOVASCULARIZATION AFTER INTRAVITREAL BEVACIZUMAB (AVASTIN) TREATMENT; ROBERT L. AVERY, MD RETINA 26:352, 2006
- 6. ELECTROPHYSIOLOGIC FINDINGS AFTER INTRAVITREAL BEVACIZUMAB (AVASTIN) TREATMENT; RAJ K. MATURI, MD, LAURA A. BLEAU, LPN, DONALD L. WILSON, MD RETINA 26:270–274, 2006
- 7. ELECTROPHYSIOLOGIC AND RETINAL PENETRATION STUDIES FOLLOWING INTRAVITREAL INJECTION OF BEVACIZUMAB (AVASTIN); JONATHAN SHAHAR, BSC,*† ROBERT L. AVERY, MD, GAD HEILWEIL, MD, ADIEL BARAK, MD, ESTHER ZEMEL, PHD, GEOFFREY P. LEWIS, PHD, PATRICK T. JOHNSON, PHD, STEVEN K. FISHER, PHD, IDO PERLMAN, PHD, ANAT LOEWENSTEIN, MD RETINA 26:262–269, 2006
- 8. INTRAVITREAL BEVACIZUMAB (AVASTIN) TREATMENT OF MACULAR EDEMA IN CENTRAL RETINAL VEIN OCCLUSION-A Short-Term Study; DIANA ITURRALDE, MD, RICHARD F. SPAIDE, MD,CATHERINE B. MEYERLE, MD, JAY M. KLANCNIK, MD,LAWRENCE A. YANNUZZI, MD, YALE L. FISHER, MD,JOHN SORENSON, MD, JASON S. SLAKTER, MD, K. BAILEY FREUND, MD,MICHAEL COONEY, MD, HOWARD F. FINE, MD, MHSC RETINA 26:279–284, 2006
- 9. INTRAVITREAL BEVACIZUMAB TREATMENT OF CHOROIDAL NEOVASCULARIZATION SECONDARY TO AGE-RELATED MACULAR DEGENERATION; RICHARD F. SPAIDE, MD, KETAN LAUD, MD,HOWARD F. FINE, MD, MHSC, JAMES M. KLANCNIK JR, MD,CATHERINE B. MEYERLE, MD, LAWRENCE A. YANNUZZI, MD,JOHN SORENSON, MD, JASON SLAKTER, MD, YALE L. FISHER, MD,MICHAEL J. COONEY, RETINA 26:383–390, 2006
- 10. Systemic Bevacizumab (Avastin) Therapy for Neovascular Age-Related Macular Degeneration: Twelve-Week Results of an Uncontrolled Open-Label Clinical Study Presented at: Macula Society Annual Meeting, February 25, 2005; Key Biscayne, Florida. Stephan Michels, MD11, Philip J. Rosenfeld, MD, PhD1 , Carmen A. Puliafito, MD, MBA1, Erin N. Marcus, MD, MPH2, Anna S. Venkatraman, MS1 Ophthalmology, Volume 112, Issue 6, Pages 1035-1047 (June 2005
- 11. Intravitreal Bevacizumab (Avastin) for Neovascular Age-Related Macular Degeneration Presented in part at: American Academy of Ophthalmology Subspecialty Meeting, October, 2005; Chicago, Illinois.Robert L. Avery, MD , Dante J. Pieramici, MD, Melvin D. Rabena, BS,Alessandro A. Castellarin, MD, Ma’an A. Nasir, MD, Matthew J. Giust, CRA
- 12. Regression of Iris Neovascularization After Intravitreal Injection of Bevacizumab in Patients With Proliferative Diabetic Retinopathy; Yusuke Oshima, MD , Hirokazu Sakaguchi, MD, Fumi Gomi, MD, Yasuo Tano, MD Ophthalmology- In Press
- 13. The international intravitreal bevacizumab safety survey: Using the internet to assess drug safety worldwide Anne E Fung , Philip J Rosenfeld and Elias Reichel . Published Online: 19 July 2006. doi:10.1136/bRITISH JOURNAL OF OPHTHALMOLOGY.2006.099598
- 14. Genentech Discloses Safety Concerns over Avastin, Nature Biotechnology 10:1198, Oct. 2004
- 15. Pharmaceutical industry sponsorship and research outcome and quality: systematic review Joel Lexchin, Lisa A Bero, Benjamin Djulbegovic, Otavio Clark: BMJ 2003;326:1167-1170 (31 May)