- Email: [email protected]
Authors respond to controversy surrounding breast cancer study
International Journal of Surgery (2005) 3, 235e239
www.int-journal-surgery.com
EDITORIAL
Authors respond to controversy surrounding breast cancer study KEYWORDS Surgery-induced angiogenesis; Mammography paradox; Bimodal relapse pattern
We published a paper on September 13, 2005 in the International Journal of Surgery that discussed our research in breast cancer.1 As our work provided a new perspective and possible explanation of a subject steeped in controversy for a decade, the Journal issued a press release, as did the home institution of the lead author. This was widely reported in the medical and lay press including interviews and comments from one or more of the authors and, for balance in reporting, contrary views from scientists or physicians were often included (see Table 1). We want to take this opportunity to discuss the resulting favourable and unfavourable comments including the letters to the editor published in this issue of the IJS and to provide our own comments on this episode. Much of the criticisms state that our paper does not prove its case. This is a very important issue as it involves the quality of the discussion from two points: the logic of our arguments and the level of understanding of these by the discussant. As for the first point, we believe that we should not be ‘‘interpreted’’ and that only our stated remarks should be attributed to us. We never argued that we were providing some ‘‘proof’’ of something. Our paper simply suggests a possible biology-based explanation of clinical observations, and no statements about ‘‘proven hypotheses’’ were reported.
Regarding the second point, we never tire of repeating what all researchers know: It is virtually impossible to prove that a given model is true, models can only be disproved when they result in significant departure from observed findings. We wish to affirm here that if our explanation will be proven unfounded (this might be done), we will reject it and we will accept a better one. This is the way by which science moves forward. Critics frequently noted that breast cancer in young women is more aggressive and that is probably the reason for the early relapses among women aged 40e49 years. The point is not to deny that this ‘‘aggressiveness’’ occurs, the point is ‘‘Why?’’ And ‘‘Why premenopausal women?’’. We suggested2e6 that ‘‘aggressiveness’’ is the putative result of surgery induced angiogenesis. Since 20% of young women with node-positive disease would relapse within one year of surgery in absence of chemotherapy, that is certainly an expression of aggressive behaviour. We suggest that this is ordinary tumour growth but kick-started out of dormancy by surgery. Furthermore we suggested that it happens among premenopausal women because of their hormone milieu4,5 and that a further factor may play a role i.e., timing of surgery in the menstrual
1743-9191/$ - see front matter ª 2005 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijsu.2005.10.017
236 Table 1
Editorial A table of the responses from researchers and clinicians published in the media
Person commenting
Publication/source
Their comments and authors reply (in italics)
Michael O’Reilly, MD Anderson Centre, Texas, USA
Wall Street Journal, 13 September 2005
Robert Smith, American Cancer Society
WebMD Health, 13 September 2005
Daniel Kopans, Massachusetts General Hospital, USA Patrick Borgen, Memorial Sloan Kettering Cancer Center, USA Susan Love, Dr. Susan Love Research Foundation, USA
Boston Globe, 13 September 2005
.animal data suggest that surgery-induced angiogenesis exists, and could be a factor in early relapse. But it still wasn’t clear whether the phenomenon worked the same way in every cancer, how to identify at-risk patients, or even whether surgery-induced angiogenesis was the only explanation. ‘‘It may explain the phenomenon in some patients but not all of them,’’ (These points are well taken. Dr. O’Reilly was lead author of the Cell 1994 paper documenting surgery induced angiogenesis in the Lewis lung model.) Don’t believe any of this. The idea that surgical interruption of the tumor bed will cause death this rapidly just does not make sense. The idea that women became worse after surgery may stem from the fact that their prognosis may have been poorer to start with since young women tend to get more aggressive cancers. Retsky et al. misread the trial data. Retsky et al.’s data are based on observations from long ago, when breast cancer screening was in its infancy. (In the editorial, we address some of the issues raised by Drs. Smith, Kopans, Borgen and Love.) If we discourage women from getting breast cancer screening, significant gains that we have made will go down the tubes. (and from Wall Street Journal). The number of cases. are too small to permit accurate analysis. The study is mostly theory and hypotheses that could scare people from getting mammograms out of fear that the treatment will kill them.
Boston Globe, 13 September 2005
SusanLoveMD.org, http://www. susanlovemd.com/ community/flashes/ news091305.htm
‘‘The idea that surgery could cause cancer to spread originated in animal studies that were first conducted about 15 years ago. Some of these studies suggested that cancer cells that have metastasized . could begin to grow once the primary tumor is removed. Others suggested that surgery might even cause the spread of cancer. But what is true in animals is not always true in humans.’’ ‘‘.it’s impossible to say that the surgery itself was responsible for the recurrence and not some other factor, such as the timing of the surgery, hormone levels, if mammography detected the cancer, or whether treatment included chemotherapy and/or radiation. Also, it is well recognized that young women tend to have tumors with more aggressive characteristics than those that appear in older women, which could influence the findings as well. .it is very critical that news stories make it clear that this particular study is pure conjecture. The last thing that I would want to hear is that women are not getting surgery to remove breast cancer tumors because they are afraid the surgery will cause the cancer to spread.’’
Blogosphere
Hazel Thornton, University of Leicester, UK
Book of Joe, 13 September 2005, http://www. bookofjoe. com/2005/09/ behindthemedspe_ 9.html Editorial in this issue of the IJS
.one of the most explosive and controversial papers to appear in the medical literature in recent years .
Letter questions decision making process and the advisability of issuing a press release (We address these serious and valid questions below).
Editorial
237
Table 1 (continued) Person commenting
Publication/source
Their comments and authors reply (in italics)
Don Berry, MD Anderson Centre, Texas, USA
See correspondence section
Tom Marshall, University of Birmingham, UK Isaac Gukas, Oncologist, Nigeria Gail Washington, FedEx executive, USA
See correspondence section
Susan Clare, Indiana University, USA
See correspondence section
P. Sooriakumaran, Royal Surrey County Hospital, UK
See correspondence section
Paper is plausible but data could also be explained by the heterogeneity of breast cancer. (Biostatistician Dr. Berry has published papers in this field and testified to US Congress on the mammography controversy. We propose that part of that black box of heterogeneity is a predictable negative effect of early interaction in a subset of node positive premenopausal women.) Need to address the Bradford Hill criteria for causation (In the editorial, we address consistency and experimental evidence e two of the most pertinent of the nine Bradford Hill criteria.) Nigerian data are discussed (We would be interested in examining these data and comparing to the Milan database.) Discussion of some AfricaneAmericans’ common acceptance of the popular ‘‘myth’’ that cancer spreads when air hits it (We agree that surgery induced angiogenesis would have all the appearances of this often-heard expression.) This letter discusses an innovative biological state of dormant cells (As with Dr. Clare, we are also interested in how to improve therapy outcome. We think that current adjuvant therapies are mainly directed toward delaying or curing patients who would have relapsed in the first peak. Questions arise such as why these therapies are not more effective [as Dr. Clare mentioned], how can we better treat patients destined to relapse in the second peak and how do these depend on surgery [including induced tumor growth] following early or late detection.) Discusses possible relevance to prostate (There is much to learn from comparisons of clinical cancer behaviour in different organs. It is especially interesting that there is also an age effect in prostate cancer.)
See correspondence section See correspondence section
cycle might modulate the tumourehost relationship and ultimately the disease outcome.7e9 In other words, we suggested a biology-based conceptual model by applying to women some facts and mechanisms incontrovertibly demonstrated in animal models, such as tumour dormancy,10e15 surgerydriven tumour growth acceleration,16e19 and angiogenesis linked to sex hormone correlations.20 Even if humans are not mice and we are well aware that what is true in animals is not always true in humans, we are definitely persuaded that we share most of our biology with other mammals and that, as ‘‘Natura non facit saltus’’, it is perfectly justified to do this. Obviously, confirmatory data should be looked for and found in breast cancer patients. Anyway, corroboration of the general correctness of our hypotheses is to be seen in many recent publications. For example, persistency of breast cancer cells in bone marrow21 or in blood22 months and even years following primary tumour surgical removal have been documented. Angio-active molecules have been observed to wax and wane within each menstrual or estral cycle.23e25 Moreover, the failure of continuous growth to explain clinical findings, such as breast cancer local recurrences, and
the need to assume some tumour growth interruption, were confirmed a few years ago.26 About the possible explanation of the mammography screening paradox in younger women that we proposed, the first point is the occurrence of the phenomenon. Some have denied the occurrence of an early mortality excess for younger women and suggest that it represents a statistically insignificant artefact. We do not have a fidelistic attitude towards the p-value from either univariate or multivariate analysis but wish to emphasize the fact that the early mortality excess of invited women is documented consistently across screening trials, countries and time, supports its clinical and scientific significance. The second point is to agree that the surge in mortality at the third year is very difficult to explain if tumours are autonomous, even by recourse to the heterogeneity of breast cancer. We state that it would not be possible unless something synchronized screened young women and initiated sudden tumour growth. Our study of the Milan breast cancer database strongly suggested that something near the time of surgery induced angiogenesis in 20% of premenopausal women with node positive disease. With that
238 assumption, we were able to calculate the result of this in a trial of early detection of breast cancer. That turned out to be 0.1 deaths per 1000 screened young women in the third year. That is the same value that is seen in the trials. This would be a temporary effect (concerning the trials e obviously not the affected individuals) and eventually it would appear that screening reduced mortality. This is also seen in the trials. We were very impressed with the meta-analysis data from Cox27 that were based on over 800,000 person-years of experience in both the intervention arm and the control arm. We called this indirect evidence but, based on our findings, that was a conservative statement and we think far stronger than pure conjecture. Most criticisms came from physicians involved in mammography screening, who stigmatized any idea of reconsidering screening guidelines for young women by stating that the paper will frighten young women into avoiding mammography or, from bad to worse, into avoiding surgery to remove breast cancer. In the paper, we called for the reconsideration of guidelines for early detection of breast cancer and suggested that, at the very least, women need to be advised of the information presented. Obviously, we were not addressing this to lay persons but to specialists and we are very surprised at their foreclosure of any possible doubt about a subject that in the past had such a wavering course. To reconsider does not mean to change, but only to take into account some other point to verify whether we are doing the right thing. In our opinion, as each man and woman is landlord of his or her own life, he or she is entitled to have all relevant elements to make decisions. The majority report and the minority report of the 1997 NIH Consensus Conference strongly disagreed on the key recommendations but both agreed that women needed to be advised of the risks and benefits of early detection. Anyway, after seeing the widespread publicity occurring, we took care to clarify that at present we are certainly not recommending any change in clinical practice but that this topic needs to be readdressed on a scientific level. We did this on our own volition. Our would-be censors are concerned that women will not understand these results and take inappropriate action. These criticisms are condescending to women implying that they are not capable of making appropriate informed decisions. We find these criticisms to be wrong but we are well aware of the technically challenging discussion and emotional climate surrounding this topic. However, we wonder, how should a scientific minority behave? Should it be silent? The minority opinion group at the 1997 NIH
Editorial Consensus Conference was not. Is science a question of votes? It is proper to address the question of the advisability of issuing press releases, mainly when the press shows the trend to sensationalize the conclusions and when errors in reporting cannot be avoided. We published papers in the past that discussed the possibility that surgery could induce angiogenesis, mostly in premenopausal patients with node positive disease, and that this would explain the mammography paradox.4,28,29 The hypotheses are testable and the risks of early detection can be eliminated with better understanding and research. But first they need to be discussed. We have been very patient waiting for this study to be debated and have not rushed into issuing press releases promoting our work. One of us (MR) is a patient advocate on the Board of Directors of the Colon Cancer Alliance. From an advocate/researcher’s perspective, the mammography paradox or controversy is odorous and needs to be properly aired out in public. This was likewise consistent with our intention to fuel a debate about the merit of the scientific proposal. We might have been wrong. Yet, to affirm with certainty that it has been so, we wish to wait for the future scientific debate among oncology researchers.
References 1. Retsky M, Demicheli R, Hrushesky WJM. Does surgery induce angiogenesis in breast cancer? Indirect evidence from relapse pattern and mammography paradox. Int J Surg Sept 2005;3:179e87. 2. Demicheli R, Retsky MW, Swartzendruber DE, Bonadonna G. Proposal for a new model of breast cancer metastatic development. Ann Oncol 1997;8:1075e80. 3. Retsky MW, Demicheli R, Swartzendruber DE, Bame PD, Wardwell RH, Bonadonna JE, et al. Computer simulation of a breast cancer metastasis model. Breast Cancer Res Treat 1997;45:193e202. 4. Retsky M, Demicheli R, Hrushesky W. Premenopausal status accelerates relapse in node positive breast cancer: hypothesis links angiogenesis, screening controversy. Breast Cancer Res Treat 2001;65:217e24. 5. Demicheli R, Bonadonna G, Hrushesky WJM, Retsky MW, Valagussa P. Menopausal status dependence of the timing of breast cancer recurrence following primary tumour surgical removal. Breast Cancer Res 2004;6:R689e96. 6. Retsky M, Bonadonna G, Demicheli R, Folkman J, Hrushesky W, Valagussa P. Hypothesis: induced angiogenesis after surgery in premenopausal node-positive breast cancer patients is a major underlying reason why adjuvant chemotherapy works particularly well for those patients. Breast Cancer Res 2004;6:R372e4. 7. Hrushesky W, Bluming A, Gruber S, Sothern R. Menstrual influence on surgical cure of breast cancer. Lancet 1989; 8669:949e52. 8. Hagen AA, Hrushesky WJ. Menstrual timing of breast cancer surgery. Am J Surg 1998;175:245e61.
Editorial 9. Love RR, Duc NB, Dihn NV, Shen T, Havighurst TZ, Allred DC. Mastectomy and oophorectomy by menstrual cycle phase in women with operable breast cancer. J Natl Cancer Inst 2002;94:662e9. 10. Fisher B, Fisher ER. Experimental evidence in support of dormant tumour cells. Science 1959;130:918e9. 11. Holmgren L, O’Reilly MS, Folkman J. Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression. Nat Med 1995;1:149e53. 12. Naumov GN, MacDonald IC, Weinmeister PM, Kerkvliet N, Nadkarni KV, Wilson SM, et al. Persistence of solitary mammary carcinoma cells in a secondary site: a possible contributor to dormancy. Cancer Res 2002;62:2162e8. 13. Goodison S, Kawai K, Hihara J, Jiang P, Yang M, Urquidi V, et al. Prolonged dormancy and site-specific growth potential of cancer cells spontaneously disseminated from non-metastatic breast tumours as revealed by labelling with green fluorescent protein. Clin Cancer Res 2003;9: 3808e14. 14. Wieder R. Insurgent micrometastases: sleeper cells and harboring the enemy. J Surg Oncol 2005 Mar 15;89(4):207e10. 15. Udagawa T, Fernandez A, Achilles EG, Folkman J, D’Amato RJ. Persistence of microscopic human cancers in mice: alterations in the angiogenic balance accompanies loss of tumor dormancy. FASEB J 2002 Sep;16(11):1361e70. 16. Simpson-Herren L, Sanford AH, Holmquist JP. Effects of surgery on the cell kinetics of residual tumour. Cancer Treat Rep 1976;60:1749e60. 17. Gunduz N, Fisher B, Saffer EA. Effect of surgical removal on the growth and kinetics of residual tumour. Cancer Res 1979;39:3861e5. 18. O’Reilly MS, Holmgren L, Shing Y, Chen C, Rosenthal RA, Moses M, et al. Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 1994;79:315e28. 19. Guba M, Cernaianu G, Koehl G, Geissler EK, Jaunch KW, Anthuber M, et al. A primary tumour promotes dormancy of solitary tumour cells before inhibiting angiogenesis. Cancer Res 2001;61:5575e9. 20. Wood PA, Bove K, You S, Chambers A, Hrushesky WJM. Cancer growth and spread are salutatory and phase-locked to the reproductive cycle through mediators of angiogenesis. Mol Cancer Ther 2005;4:1065e75. 21. Pantel K, Otte M. Occult micrometastasis: enrichment, identification and characterization of single disseminated tumour cells. Semin Cancer Biol 2001;11:327e37.
239 22. Meng S, Tripathy D, Frenkel PE, Shete S, Naftalis EZ, Huth JF, et al. Circulating tumour cells in patients with breast cancer dormancy. Clin Cancer Res 2004;10:8152e62. 23. Dabrosin C. Variability of vascular endothelial growth factor in normal human breast tissue in vivo during the menstrual cycle. J Clin Endocrinol Metab 2003;88: 2695e8. 24. Malamitsi-Puchner A, Tziotis J, Tsonou A, Sarandakou A, Bartsocas CS, Cretsas G. Serum angiogenin levels in the female from birth to postmenopause. Growth Factors 1999;17:75e9. 25. Kuroi K, Tanaka C, Toi M. Circulating levels of endostatin in cancer patients. Oncol Rep 2001;8:405e9. 26. Demicheli R, Terenziani M, Valagussa P, Moliterni A, Zambetti, Bonadonna G. Local recurrences following mastectomy: support for the concept of tumour dormancy. J Natl Cancer Inst 1994;86:45e8. 27. Cox B. Variation in the effectiveness of breast screening by year of follow-up. J Natl Cancer Inst Monogr 1997;22: 69e72. 28. Demicheli R, Bonadonna G, Hrushesky WJ, Retsky MW, Valagussa P. Menopausal status dependence of early mortality reduction due to diagnosis of smaller breast cancers (T1 vs T2eT3): relevance to screening. J Clin Oncol 2004;22: 102e7. 29. Retsky M, Demicheli R, Hrushesky W. Breast cancer screening: controversies and future directions. Curr Opin Obstetr Gynecol 2003;15:1e8.
Michael Retsky* Children’s Hospital/Harvard Medical School, Vascular Biology, Karp Family Laboratories, 12th Floor, 1 Blackfan Circle, Boston, MA 02115, USA Romano Demicheli Istituto Nazionale Tumori, Milano, Italy William J.M. Hrushesky WJB Dorn Veterans Affairs Medical Centre, Columbia, USA *Corresponding author. Tel.: þ1 11 203 452 1649. E-mail address: michael.retsky@childrens. harvard.edu
www.int-journal-surgery.com
EDITORIAL
Authors respond to controversy surrounding breast cancer study KEYWORDS Surgery-induced angiogenesis; Mammography paradox; Bimodal relapse pattern
We published a paper on September 13, 2005 in the International Journal of Surgery that discussed our research in breast cancer.1 As our work provided a new perspective and possible explanation of a subject steeped in controversy for a decade, the Journal issued a press release, as did the home institution of the lead author. This was widely reported in the medical and lay press including interviews and comments from one or more of the authors and, for balance in reporting, contrary views from scientists or physicians were often included (see Table 1). We want to take this opportunity to discuss the resulting favourable and unfavourable comments including the letters to the editor published in this issue of the IJS and to provide our own comments on this episode. Much of the criticisms state that our paper does not prove its case. This is a very important issue as it involves the quality of the discussion from two points: the logic of our arguments and the level of understanding of these by the discussant. As for the first point, we believe that we should not be ‘‘interpreted’’ and that only our stated remarks should be attributed to us. We never argued that we were providing some ‘‘proof’’ of something. Our paper simply suggests a possible biology-based explanation of clinical observations, and no statements about ‘‘proven hypotheses’’ were reported.
Regarding the second point, we never tire of repeating what all researchers know: It is virtually impossible to prove that a given model is true, models can only be disproved when they result in significant departure from observed findings. We wish to affirm here that if our explanation will be proven unfounded (this might be done), we will reject it and we will accept a better one. This is the way by which science moves forward. Critics frequently noted that breast cancer in young women is more aggressive and that is probably the reason for the early relapses among women aged 40e49 years. The point is not to deny that this ‘‘aggressiveness’’ occurs, the point is ‘‘Why?’’ And ‘‘Why premenopausal women?’’. We suggested2e6 that ‘‘aggressiveness’’ is the putative result of surgery induced angiogenesis. Since 20% of young women with node-positive disease would relapse within one year of surgery in absence of chemotherapy, that is certainly an expression of aggressive behaviour. We suggest that this is ordinary tumour growth but kick-started out of dormancy by surgery. Furthermore we suggested that it happens among premenopausal women because of their hormone milieu4,5 and that a further factor may play a role i.e., timing of surgery in the menstrual
1743-9191/$ - see front matter ª 2005 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijsu.2005.10.017
236 Table 1
Editorial A table of the responses from researchers and clinicians published in the media
Person commenting
Publication/source
Their comments and authors reply (in italics)
Michael O’Reilly, MD Anderson Centre, Texas, USA
Wall Street Journal, 13 September 2005
Robert Smith, American Cancer Society
WebMD Health, 13 September 2005
Daniel Kopans, Massachusetts General Hospital, USA Patrick Borgen, Memorial Sloan Kettering Cancer Center, USA Susan Love, Dr. Susan Love Research Foundation, USA
Boston Globe, 13 September 2005
.animal data suggest that surgery-induced angiogenesis exists, and could be a factor in early relapse. But it still wasn’t clear whether the phenomenon worked the same way in every cancer, how to identify at-risk patients, or even whether surgery-induced angiogenesis was the only explanation. ‘‘It may explain the phenomenon in some patients but not all of them,’’ (These points are well taken. Dr. O’Reilly was lead author of the Cell 1994 paper documenting surgery induced angiogenesis in the Lewis lung model.) Don’t believe any of this. The idea that surgical interruption of the tumor bed will cause death this rapidly just does not make sense. The idea that women became worse after surgery may stem from the fact that their prognosis may have been poorer to start with since young women tend to get more aggressive cancers. Retsky et al. misread the trial data. Retsky et al.’s data are based on observations from long ago, when breast cancer screening was in its infancy. (In the editorial, we address some of the issues raised by Drs. Smith, Kopans, Borgen and Love.) If we discourage women from getting breast cancer screening, significant gains that we have made will go down the tubes. (and from Wall Street Journal). The number of cases. are too small to permit accurate analysis. The study is mostly theory and hypotheses that could scare people from getting mammograms out of fear that the treatment will kill them.
Boston Globe, 13 September 2005
SusanLoveMD.org, http://www. susanlovemd.com/ community/flashes/ news091305.htm
‘‘The idea that surgery could cause cancer to spread originated in animal studies that were first conducted about 15 years ago. Some of these studies suggested that cancer cells that have metastasized . could begin to grow once the primary tumor is removed. Others suggested that surgery might even cause the spread of cancer. But what is true in animals is not always true in humans.’’ ‘‘.it’s impossible to say that the surgery itself was responsible for the recurrence and not some other factor, such as the timing of the surgery, hormone levels, if mammography detected the cancer, or whether treatment included chemotherapy and/or radiation. Also, it is well recognized that young women tend to have tumors with more aggressive characteristics than those that appear in older women, which could influence the findings as well. .it is very critical that news stories make it clear that this particular study is pure conjecture. The last thing that I would want to hear is that women are not getting surgery to remove breast cancer tumors because they are afraid the surgery will cause the cancer to spread.’’
Blogosphere
Hazel Thornton, University of Leicester, UK
Book of Joe, 13 September 2005, http://www. bookofjoe. com/2005/09/ behindthemedspe_ 9.html Editorial in this issue of the IJS
.one of the most explosive and controversial papers to appear in the medical literature in recent years .
Letter questions decision making process and the advisability of issuing a press release (We address these serious and valid questions below).
Editorial
237
Table 1 (continued) Person commenting
Publication/source
Their comments and authors reply (in italics)
Don Berry, MD Anderson Centre, Texas, USA
See correspondence section
Tom Marshall, University of Birmingham, UK Isaac Gukas, Oncologist, Nigeria Gail Washington, FedEx executive, USA
See correspondence section
Susan Clare, Indiana University, USA
See correspondence section
P. Sooriakumaran, Royal Surrey County Hospital, UK
See correspondence section
Paper is plausible but data could also be explained by the heterogeneity of breast cancer. (Biostatistician Dr. Berry has published papers in this field and testified to US Congress on the mammography controversy. We propose that part of that black box of heterogeneity is a predictable negative effect of early interaction in a subset of node positive premenopausal women.) Need to address the Bradford Hill criteria for causation (In the editorial, we address consistency and experimental evidence e two of the most pertinent of the nine Bradford Hill criteria.) Nigerian data are discussed (We would be interested in examining these data and comparing to the Milan database.) Discussion of some AfricaneAmericans’ common acceptance of the popular ‘‘myth’’ that cancer spreads when air hits it (We agree that surgery induced angiogenesis would have all the appearances of this often-heard expression.) This letter discusses an innovative biological state of dormant cells (As with Dr. Clare, we are also interested in how to improve therapy outcome. We think that current adjuvant therapies are mainly directed toward delaying or curing patients who would have relapsed in the first peak. Questions arise such as why these therapies are not more effective [as Dr. Clare mentioned], how can we better treat patients destined to relapse in the second peak and how do these depend on surgery [including induced tumor growth] following early or late detection.) Discusses possible relevance to prostate (There is much to learn from comparisons of clinical cancer behaviour in different organs. It is especially interesting that there is also an age effect in prostate cancer.)
See correspondence section See correspondence section
cycle might modulate the tumourehost relationship and ultimately the disease outcome.7e9 In other words, we suggested a biology-based conceptual model by applying to women some facts and mechanisms incontrovertibly demonstrated in animal models, such as tumour dormancy,10e15 surgerydriven tumour growth acceleration,16e19 and angiogenesis linked to sex hormone correlations.20 Even if humans are not mice and we are well aware that what is true in animals is not always true in humans, we are definitely persuaded that we share most of our biology with other mammals and that, as ‘‘Natura non facit saltus’’, it is perfectly justified to do this. Obviously, confirmatory data should be looked for and found in breast cancer patients. Anyway, corroboration of the general correctness of our hypotheses is to be seen in many recent publications. For example, persistency of breast cancer cells in bone marrow21 or in blood22 months and even years following primary tumour surgical removal have been documented. Angio-active molecules have been observed to wax and wane within each menstrual or estral cycle.23e25 Moreover, the failure of continuous growth to explain clinical findings, such as breast cancer local recurrences, and
the need to assume some tumour growth interruption, were confirmed a few years ago.26 About the possible explanation of the mammography screening paradox in younger women that we proposed, the first point is the occurrence of the phenomenon. Some have denied the occurrence of an early mortality excess for younger women and suggest that it represents a statistically insignificant artefact. We do not have a fidelistic attitude towards the p-value from either univariate or multivariate analysis but wish to emphasize the fact that the early mortality excess of invited women is documented consistently across screening trials, countries and time, supports its clinical and scientific significance. The second point is to agree that the surge in mortality at the third year is very difficult to explain if tumours are autonomous, even by recourse to the heterogeneity of breast cancer. We state that it would not be possible unless something synchronized screened young women and initiated sudden tumour growth. Our study of the Milan breast cancer database strongly suggested that something near the time of surgery induced angiogenesis in 20% of premenopausal women with node positive disease. With that
238 assumption, we were able to calculate the result of this in a trial of early detection of breast cancer. That turned out to be 0.1 deaths per 1000 screened young women in the third year. That is the same value that is seen in the trials. This would be a temporary effect (concerning the trials e obviously not the affected individuals) and eventually it would appear that screening reduced mortality. This is also seen in the trials. We were very impressed with the meta-analysis data from Cox27 that were based on over 800,000 person-years of experience in both the intervention arm and the control arm. We called this indirect evidence but, based on our findings, that was a conservative statement and we think far stronger than pure conjecture. Most criticisms came from physicians involved in mammography screening, who stigmatized any idea of reconsidering screening guidelines for young women by stating that the paper will frighten young women into avoiding mammography or, from bad to worse, into avoiding surgery to remove breast cancer. In the paper, we called for the reconsideration of guidelines for early detection of breast cancer and suggested that, at the very least, women need to be advised of the information presented. Obviously, we were not addressing this to lay persons but to specialists and we are very surprised at their foreclosure of any possible doubt about a subject that in the past had such a wavering course. To reconsider does not mean to change, but only to take into account some other point to verify whether we are doing the right thing. In our opinion, as each man and woman is landlord of his or her own life, he or she is entitled to have all relevant elements to make decisions. The majority report and the minority report of the 1997 NIH Consensus Conference strongly disagreed on the key recommendations but both agreed that women needed to be advised of the risks and benefits of early detection. Anyway, after seeing the widespread publicity occurring, we took care to clarify that at present we are certainly not recommending any change in clinical practice but that this topic needs to be readdressed on a scientific level. We did this on our own volition. Our would-be censors are concerned that women will not understand these results and take inappropriate action. These criticisms are condescending to women implying that they are not capable of making appropriate informed decisions. We find these criticisms to be wrong but we are well aware of the technically challenging discussion and emotional climate surrounding this topic. However, we wonder, how should a scientific minority behave? Should it be silent? The minority opinion group at the 1997 NIH
Editorial Consensus Conference was not. Is science a question of votes? It is proper to address the question of the advisability of issuing press releases, mainly when the press shows the trend to sensationalize the conclusions and when errors in reporting cannot be avoided. We published papers in the past that discussed the possibility that surgery could induce angiogenesis, mostly in premenopausal patients with node positive disease, and that this would explain the mammography paradox.4,28,29 The hypotheses are testable and the risks of early detection can be eliminated with better understanding and research. But first they need to be discussed. We have been very patient waiting for this study to be debated and have not rushed into issuing press releases promoting our work. One of us (MR) is a patient advocate on the Board of Directors of the Colon Cancer Alliance. From an advocate/researcher’s perspective, the mammography paradox or controversy is odorous and needs to be properly aired out in public. This was likewise consistent with our intention to fuel a debate about the merit of the scientific proposal. We might have been wrong. Yet, to affirm with certainty that it has been so, we wish to wait for the future scientific debate among oncology researchers.
References 1. Retsky M, Demicheli R, Hrushesky WJM. Does surgery induce angiogenesis in breast cancer? Indirect evidence from relapse pattern and mammography paradox. Int J Surg Sept 2005;3:179e87. 2. Demicheli R, Retsky MW, Swartzendruber DE, Bonadonna G. Proposal for a new model of breast cancer metastatic development. Ann Oncol 1997;8:1075e80. 3. Retsky MW, Demicheli R, Swartzendruber DE, Bame PD, Wardwell RH, Bonadonna JE, et al. Computer simulation of a breast cancer metastasis model. Breast Cancer Res Treat 1997;45:193e202. 4. Retsky M, Demicheli R, Hrushesky W. Premenopausal status accelerates relapse in node positive breast cancer: hypothesis links angiogenesis, screening controversy. Breast Cancer Res Treat 2001;65:217e24. 5. Demicheli R, Bonadonna G, Hrushesky WJM, Retsky MW, Valagussa P. Menopausal status dependence of the timing of breast cancer recurrence following primary tumour surgical removal. Breast Cancer Res 2004;6:R689e96. 6. Retsky M, Bonadonna G, Demicheli R, Folkman J, Hrushesky W, Valagussa P. Hypothesis: induced angiogenesis after surgery in premenopausal node-positive breast cancer patients is a major underlying reason why adjuvant chemotherapy works particularly well for those patients. Breast Cancer Res 2004;6:R372e4. 7. Hrushesky W, Bluming A, Gruber S, Sothern R. Menstrual influence on surgical cure of breast cancer. Lancet 1989; 8669:949e52. 8. Hagen AA, Hrushesky WJ. Menstrual timing of breast cancer surgery. Am J Surg 1998;175:245e61.
Editorial 9. Love RR, Duc NB, Dihn NV, Shen T, Havighurst TZ, Allred DC. Mastectomy and oophorectomy by menstrual cycle phase in women with operable breast cancer. J Natl Cancer Inst 2002;94:662e9. 10. Fisher B, Fisher ER. Experimental evidence in support of dormant tumour cells. Science 1959;130:918e9. 11. Holmgren L, O’Reilly MS, Folkman J. Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression. Nat Med 1995;1:149e53. 12. Naumov GN, MacDonald IC, Weinmeister PM, Kerkvliet N, Nadkarni KV, Wilson SM, et al. Persistence of solitary mammary carcinoma cells in a secondary site: a possible contributor to dormancy. Cancer Res 2002;62:2162e8. 13. Goodison S, Kawai K, Hihara J, Jiang P, Yang M, Urquidi V, et al. Prolonged dormancy and site-specific growth potential of cancer cells spontaneously disseminated from non-metastatic breast tumours as revealed by labelling with green fluorescent protein. Clin Cancer Res 2003;9: 3808e14. 14. Wieder R. Insurgent micrometastases: sleeper cells and harboring the enemy. J Surg Oncol 2005 Mar 15;89(4):207e10. 15. Udagawa T, Fernandez A, Achilles EG, Folkman J, D’Amato RJ. Persistence of microscopic human cancers in mice: alterations in the angiogenic balance accompanies loss of tumor dormancy. FASEB J 2002 Sep;16(11):1361e70. 16. Simpson-Herren L, Sanford AH, Holmquist JP. Effects of surgery on the cell kinetics of residual tumour. Cancer Treat Rep 1976;60:1749e60. 17. Gunduz N, Fisher B, Saffer EA. Effect of surgical removal on the growth and kinetics of residual tumour. Cancer Res 1979;39:3861e5. 18. O’Reilly MS, Holmgren L, Shing Y, Chen C, Rosenthal RA, Moses M, et al. Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 1994;79:315e28. 19. Guba M, Cernaianu G, Koehl G, Geissler EK, Jaunch KW, Anthuber M, et al. A primary tumour promotes dormancy of solitary tumour cells before inhibiting angiogenesis. Cancer Res 2001;61:5575e9. 20. Wood PA, Bove K, You S, Chambers A, Hrushesky WJM. Cancer growth and spread are salutatory and phase-locked to the reproductive cycle through mediators of angiogenesis. Mol Cancer Ther 2005;4:1065e75. 21. Pantel K, Otte M. Occult micrometastasis: enrichment, identification and characterization of single disseminated tumour cells. Semin Cancer Biol 2001;11:327e37.
239 22. Meng S, Tripathy D, Frenkel PE, Shete S, Naftalis EZ, Huth JF, et al. Circulating tumour cells in patients with breast cancer dormancy. Clin Cancer Res 2004;10:8152e62. 23. Dabrosin C. Variability of vascular endothelial growth factor in normal human breast tissue in vivo during the menstrual cycle. J Clin Endocrinol Metab 2003;88: 2695e8. 24. Malamitsi-Puchner A, Tziotis J, Tsonou A, Sarandakou A, Bartsocas CS, Cretsas G. Serum angiogenin levels in the female from birth to postmenopause. Growth Factors 1999;17:75e9. 25. Kuroi K, Tanaka C, Toi M. Circulating levels of endostatin in cancer patients. Oncol Rep 2001;8:405e9. 26. Demicheli R, Terenziani M, Valagussa P, Moliterni A, Zambetti, Bonadonna G. Local recurrences following mastectomy: support for the concept of tumour dormancy. J Natl Cancer Inst 1994;86:45e8. 27. Cox B. Variation in the effectiveness of breast screening by year of follow-up. J Natl Cancer Inst Monogr 1997;22: 69e72. 28. Demicheli R, Bonadonna G, Hrushesky WJ, Retsky MW, Valagussa P. Menopausal status dependence of early mortality reduction due to diagnosis of smaller breast cancers (T1 vs T2eT3): relevance to screening. J Clin Oncol 2004;22: 102e7. 29. Retsky M, Demicheli R, Hrushesky W. Breast cancer screening: controversies and future directions. Curr Opin Obstetr Gynecol 2003;15:1e8.
Michael Retsky* Children’s Hospital/Harvard Medical School, Vascular Biology, Karp Family Laboratories, 12th Floor, 1 Blackfan Circle, Boston, MA 02115, USA Romano Demicheli Istituto Nazionale Tumori, Milano, Italy William J.M. Hrushesky WJB Dorn Veterans Affairs Medical Centre, Columbia, USA *Corresponding author. Tel.: þ1 11 203 452 1649. E-mail address: michael.retsky@childrens. harvard.edu