Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial

PART 1:

This commentary addresses several points Jacob Levman had brought forward on April 7th [1] pertaining to my previous letter [2].

For example, Levman paints a mistaken and inexcusably misleading picture of Peter C. Gotzsche's stance on mammography, specifically in relation to mortality. Levman writes [1] "Gøtzsche concedes that mortality benefits are obtainable from mammographic screening", citing these two papers: [3,4]. In the first citation [3], Gøtzsche stated that the reductions in breast cancer mortality found in some studies were due to pro-screening biases or flaws in study designs and that in their absence "screening appeared ineffective".

In the second citation [4], Gøtzsche and collaborators wrote, "The observed decline in breast cancer mortality in many countries seems to be caused by improved adjuvant therapy and breast cancer awareness, not screening." - a conclusion corroborated by other investigators [5]. This Gøtzsche et al. paper is also the study Levman is referring to when he falsely accuses me of having stated a contradiction when Levman said this study "indicates that “the trials suggest only a 12% reduction in breast cancer mortality” in contradiction to Mr. Hefti’s statement of no mortality effect." I've never said "no mortality effect", I essentially echoed what Gøtzsche and colleagues had found (no relevant mortality benefit from screening, or: yes a mortality benefit but not from screening - see their aforementioned quote), that there was no statistically significant reduction in breast cancer mortality from longterm screening [2].

Anyone who carefully reviews Gøtzsche's overall mammography work, which considers both randomized controlled studies, rigorous observational studies, and other related research, will find statements that for over a decade have quite consistently and very clearly pointed in only one direction, such as “Screening for breast cancer with mammography is unjustified.” [6], "no reliable evidence shows that breast screening saves lives" [7], "Screening seems ineffective [...]" [8] and, "It is time to stop screening. This is my scientific conclusion [...]" [9] or, if still in doubt, read his book [10].

It is also strongly misleading for Levman to claim that "Two of Gøtzsche’s three adequate trials reported a mortality benefit from mammographic breast cancer screening (the UK Age trial and Malmo I) [...]" [1] when the UK Age trial [11] showed no statistically significant reduction in mortality from breast cancer and the Malmo I [12] trial found "no overall effect on the mortality from breast cancer" and concluded that "The results of our study cannot be used to advocate introduction of mammographic screening in all ages in an urban population" [13,14].

Levman pointed out that I rely heavily on Gøtzsche's work [1]. Yes, admittedly, I do rely heavily on Gøtzsche's and his allied Cochrane collaboration's work. Presumably anyone else would too if they were to recognize that their mammogram research is among the least biased, most transparent, and most substantial works available to the public (Gøtzsche's been one of the most prolific mammogram scientists ever), rather unlike many of the pro-mammogram investigations that are shrouded in secrecy about their raw data, encrusted with complicated statistical models, or heavily afflicted by conflict-of-interests. Historically, secrecy, complexity, and vested interests mean one thing: control of a specific ideology by a certain group of people, whether political or medical.

Because the mortality rates of breast cancer are low without screening, few women will benefit from screening [15]. In juxtaposition, regarding the alleged "significance" of lethal cancers found in the prevalence screening over breast examination in the Miller team trial [16], the prevalence of breast cancer in women of screening age is under 1% [17] and over 99% of screening participants will be healthy, both before and after screening, making this a very small absolute gain. And if you add my point from my previous letter [2] and the fact that some screen-detected invasive ("lethal") breast cancers can grow so slow they'd be cases of overdiagnosis or end in spontaneous remission [9,18,19], the "relative significance" becomes even more an "absolute insignificance" (and some participants may get needlessly harmed).

Levman claim (regarding the detection of late-stage cancer in the prevalence screening) that "Mortality will not improve from the detection of disease that has already reached a lethal stage of development" [1] has to be understood and viewed from his belief in the doctrinaire but erroneous "standard model of malignant tumour progression" [1] which assumes breast cancers grow linearly and progressively from a benign, non-lethal, early-stage to an invasive, advanced, lethal, malignant stage. Yet there is substantial scientific data showing that even advanced, malignant, "lethal" cancers can spontaneously regress or grow exceedingly slow as to constitute cases of overdiagnosis [9,18,19]. In fact, mortality could actually increase in mammography-detected advanced lethal cancers because an overdiagnosed and overtreated patient with a so-called "lethal" cancer, for instance, only receives harm, possibly leading to the premature demise of the patient from the aggressive interventions.

Levman remarked that the Bleyer & Welch study [5] assumed "that every in situ cancer detected by mammography is a case of overdiagnosis" - an assumption he calls "absurd" [1]. First off, Bleyer & Welch did only include ductal carcinoma in situ (DCIS) cancers while lobular carcinoma in situ cancers were excluded from their study [5]. Secondly, how much less "absurd" is their assumption really when over 50% of carcinoma in situ (CIS) cancers represent cases of overdiagnosis [10] and that "practically all women diagnosed with CIS are treated as if the condition would progress to invasive cancer, which leads to considerable overtreatment" [15]? Levman's "absurdity" clause on the extent of in situ overdiagnosis unintentionally but misleadingly deflects from an extremely disconcerting reality.

Levman's casual dismissal [1] of "Baum's work" [20] is also erroneously leading the reader into thinking the issue of harm from radiation therapy is of no or inconsequential relevance. This significant source of injury had long been reported by various investigations [21,22], but typically has been disregarded by pro-mammogram supporters in their estimates of benefits of mammography, thus "radiation treatments" had not been systematically replaced by "alternative therapies". Early-stage breast cancers still get commonly treated by radiation therapy [9]. Thus, contrary to Levman's accusation, I'm not "tarnishing mammography with the deficiencies of different technology", instead he is "tarnishing" the real facts of the matter. Furthermore, other "alternative therapies", such as chemotherapy, also increase the risk of death in overdiagnosed individuals [9].

(continues with PART 2)

References (see PART 2)