Acquired Gefitinib-Resistant Mutation of EGFR in a Chemonaive Lung Adenocarcinoma Harboring Gefitinib-Sensitive Mutation L858R

Steve Hickey, Hilary Roberts The Cochrane review by Douglas et al. [1], which is referenced in the Best Practice article by Douglas and Hemilä [2], covers 60 years of research into vitamin C and the common cold. However, the review omits pharmacokinetic data that invalidate the conclusion that vitamin C is ineffective. This conclusion is not derivable from the data presented. The dual-phase pharmacokinetics of vitamin C are described by the dynamic fl ow model [3,4]. Low gram-level intakes of ascorbate, leading to blood plasma levels below 70 μM/l, have a half-life of 8–40 days. Higher gram-level intakes have a plasma half-life of 30 minutes [3]. A large oral dose raises blood plasma levels briefl y: they reach a peak after two to three hours, before decaying back to baseline. Frequent repeated doses allow sustained high plasma levels of about 250 μM/l [4,5]. Douglas and Hemilä reviewed intakes that transiently raise plasma ascorbate levels above 70 μM/l. A single dose does not raise the median level [6,7]. Daily supplements would, thus, not increase disease resistance to any great degree [3,4]. Single or double doses daily will not increase background plasma levels, regardless of the magnitude of the dose [6,7]. Since plasma ascorbate is at background level for the majority of the day, effects will be minimal. There is widespread confusion about nutritional and pharmacological levels of supplementation [3]. Linus Pauling, typically, described nutritional gram-level doses able to provide a degree of disease prevention [8]. By contrast, pharmacological doses used for treatment are, at minimum, an order of magnitude larger and involve frequent doses. The doses should be at intervals of three hours or less [3]. Treatment doses are described by Cathcart’s paper on titration to bowel tolerance [9]. To treat the onset of a cold, the therapy is perhaps a minimum of 10 g of oral ascorbic acid, followed by at least 2 g each hour [3,4]. Douglas and Hemilä give a misleading impression by not making it clear that the doses they consider are not pharmacological. They claim that the results of one study, giving an 8-g dose at the start of symptoms, are tantalising and deserve further assessment. However, once this single dose has been excreted, the protective effects will be lost. During illness, ascorbate is depleted rapidly and higher oral intakes are tolerated—up to 200 g per day [9]. It would be surprising if this 8-g dose had a large effect. Studies on ascorbate require appropriate doses. Douglas and Hemilä have only confi rmed that 60 years of vitamin C research has largely been wasted because of confusion between nutritional and pharmacological intakes, and because of a misunderstanding of the pharmacokinetics. It is essential that high-dose studies take into account ascorbate’s dual-phase pharmacokinetics. The dosing regime should allow sustained high plasma levels to be achieved. The claim that vitamin C cannot prevent or cure the common cold is both premature and unwarranted. Steve Hickey (radicalascorbate@yahoo.com)


Misleading Information on the Properties of Vitamin C Steve Hickey, Hilary Roberts
The Cochrane review by Douglas et al. [1], which is referenced in the Best Practice article by Douglas and Hemilä [2], covers 60 years of research into vitamin C and the common cold.However, the review omits pharmacokinetic data that invalidate the conclusion that vitamin C is ineffective.This conclusion is not derivable from the data presented.
The dual-phase pharmacokinetics of vitamin C are described by the dynamic fl ow model [3,4].Low gram-level intakes of ascorbate, leading to blood plasma levels below 70 µM/l, have a half-life of 8-40 days.Higher gram-level intakes have a plasma half-life of 30 minutes [3].A large oral dose raises blood plasma levels briefl y: they reach a peak after two to three hours, before decaying back to baseline.Frequent repeated doses allow sustained high plasma levels of about 250 µM/l [4,5].
Douglas and Hemilä reviewed intakes that transiently raise plasma ascorbate levels above 70 µM/l.A single dose does not raise the median level [6,7].Daily supplements would, thus, not increase disease resistance to any great degree [3,4].Single or double doses daily will not increase background plasma levels, regardless of the magnitude of the dose [6,7].Since plasma ascorbate is at background level for the majority of the day, effects will be minimal.
There is widespread confusion about nutritional and pharmacological levels of supplementation [3].Linus Pauling, typically, described nutritional gram-level doses able to provide a degree of disease prevention [8].By contrast, pharmacological doses used for treatment are, at minimum, an order of magnitude larger and involve frequent doses.The doses should be at intervals of three hours or less [3].Treatment doses are described by Cathcart's paper on titration to bowel tolerance [9].To treat the onset of a cold, the therapy is perhaps a minimum of 10 g of oral ascorbic acid, followed by at least 2 g each hour [3,4].
Douglas and Hemilä give a misleading impression by not making it clear that the doses they consider are not pharmacological.They claim that the results of one study, giving an 8-g dose at the start of symptoms, are tantalising and deserve further assessment.However, once this single dose has been excreted, the protective effects will be lost.During illness, ascorbate is depleted rapidly and higher oral intakes are tolerated-up to 200 g per day [9].It would be surprising if this 8-g dose had a large effect.
Studies on ascorbate require appropriate doses.Douglas and Hemilä have only confi rmed that 60 years of vitamin C research has largely been wasted because of confusion between nutritional and pharmacological intakes, and because of a misunderstanding of the pharmacokinetics.It is essential that high-dose studies take into account ascorbate's dual-phase pharmacokinetics.The dosing regime should allow sustained high plasma levels to be achieved.The claim that vitamin C cannot prevent or cure the common cold is both premature and unwarranted.

Authors' Reply
The responses to our Best Practice article [1] by Hickey and Roberts [2], and by Sardi [3], make the same point, namely, that a recent pharmacokinetic study reported that frequent oral intakes of vitamin C would be necessary to elevate plasma ascorbic acid levels to the point where they believe it would have a pharmacological impact.Both authors suggest that the conclusions of our Cochrane review [4] are fl awed because all of the placebo-controlled trials that have been carried out so far have used, for both prophylaxis and therapy, one to three doses per day of vitamin C, ranging from 200 mg daily to as much as 8 g in a single daily dose.
We have not, as our critics imply, concluded that vitamin C, in the doses used in trials reported in the literature, has no effect on the common cold.On the contrary, our evidence indicated that in marathon runners and in those exposed to high physical or cold stress, a substantial prophylactic effect was observed.And in the general population using regular vitamin C prophylaxis, common cold duration was consistently shortened, but the level of shortening was relatively trivial.
We do not consider the vitamin C and the common cold relationship closed.Nor are we persuaded by the arguments of these three critics that frequent, large doses would necessarily result in substantially greater benefi ts than earlier trials have demonstrated.
We consider that it may be useful to distinguish between (a) prophylactic supplementation for people who are in good health and (b) therapeutic supplementation for people who have an infection.The kidneys reabsorb essentially all vitamin C when the dietary intake is below 60-100 mg/day, and the vitamin C level in leukocytes is saturated by approximately 100 mg/day [5]; in this respect, we doubt that prophylactic supplementation of healthy people, using doses higher than those in the published trials, might be expected to benefi t the general healthy population.On the other hand, there is evidence indicating that common cold infection decreases the vitamin C level in leukocytes, suggesting changes in vitamin C metabolism [6], and, in this respect, there seems to be a rationale to study the effects of supplementation on people infected with the common cold using even higher doses.
To this point, the claim that these two letters make has not been reported in properly conducted randomized controlled trials of either therapy or prophylaxis.We look forward to incorporating such trials, when they have been carried out, in future versions of the Cochrane review.Meanwhile, we stand fi rmly by the conclusions reported in our article.

Dave Hambidge
May I respectfully highlight a potential confounding factor to interpreting an otherwise excellent and provoking study by Saha et al. [1].
In their recent overview of secondary schizophrenia (defi ned as "a disparate range of brain disorders that can, uncommonly, give rise to schizophrenia like symptomology" [2]), Hyde and Lewis [2] concluded that, overall, there was a prevalence rate of 5%-8% for psychoses of likely identifi able organic etiology amongst a series of relatively unselected patients.They suggest screening procedures in new cases of psychosis, including schizophrenia, with a battery of blood tests, a urine drug screen (UDS), and an electroencephalogram (EEG) as fi rst-line investigations.
Between September 2000 and November 2003, I interviewed and studied the medical records of 56 patients in northwest England, who were appealing against detention under the Mental Health Act (1983), and who had been admitted for the fi rst time within the last ten years [3].They were all referred to me by their solicitors to prepare Legal Aid/Legal Services Commission-funded independent reports for their Mental Health Review Tribunal hearings.For each patient, I recorded which of the organic investigations suggested by Hyde and Lewis, if any, had been undertaken.
Of the 56 patients, ten were being detained for the fi rst time (three females and seven males, detained on average for 39 weeks) and 13 had been detained for over one year (two females and 11 males, detained on average for 106 weeks).Whilst all except two of the 56 patients had some combination of blood tests recorded, 55% did not have a UDS and 83% did not have an EEG.Syphilis serology was examined for in only two patients of the latter group and none of the former.Therefore, my fi ndings suggest that secondary schizophrenias may not be investigated for in most detained patients with a schizophrenia-like illness in England.
As secondary schizophrenias are present in 5%-8% of such cases, some of the variability in rates found by these authors must be related to the differing diagnostic rigour used to exclude secondary causes.

Authors' Reply: Measurement Errors in Schizophrenia Epidemiology
The letter from Hambidge highlights the heterogeneous nature of schizophrenia [1].In order to diagnose schizophrenia, modern diagnostic criteria require the exclusion of other general somatic conditions that can mimic psychotic symptoms.Compliance with screening protocols designed to identify these disorders varies widely, even in developed countries.We agree with the correspondent that some studies included in our recent systematic review [2] would have probably included individuals who were subsequently found to have "secondary schizophrenia" (i.e., false positives).Thus, this issue would slightly infl ate the prevalence estimate.The inappropriate inclusion of false positives is only one of a very long list of methodological factors that contribute to imprecision in the estimation of the incidence and prevalence of schizophrenia.The critical issue for the research community is how best to partition out measurement error from "true" variations in the incidence or prevalence of schizophrenia.In the absence of more refi ned phenotypes for the many different disorders that contribute to the syndrome of schizophrenia (e.g., by the use of yet-tobe-identifi ed biomarkers), standard epidemiological studies of the incidence and prevalence of schizophrenia may have reached their limits of precision.Stampfer's recent Perspective [1] on the paper by Sørensen et al. [2] appropriately acknowledges the challenges inherent in using observational epidemiology to determine the impact of weight loss on life expectancy.However, his case that the data of Sørensen et al. do not support their conclusion that intentional weight loss may be hazardous is based, in part, on erroneous statements about the study.
Stampfer suggests that "reverse causation" could account for the fi ndings of Sørensen et al. because he believes they did not do a "lagged" analysis in which deaths that occur in the fi rst few years after follow-up are excluded.In the statistical analysis, however, Sørensen et al. describe using two separate fully adjusted models: one for the fi rst fi ve years of follow-up and one for the period thereafter, and they also reported mortality hazard ratios (HRs) associated with intentional weight loss during each period.Because so few deaths occurred in the fi rst fi ve years of follow-up, the estimated mortality HR for intentional weight loss during this period (6.26) had such a wide confi dence interval (0.33-118) that it was essentially meaningless.However, after excluding the fi rst fi ve years of follow-up data, Sørensen et al. still found a clinically and statistically signifi cant association between intentional weight loss and death during the remaining 13 years of follow-up: HR = 1.88 (confi dence interval, 1.05-3.39).
Stampfer indicates that the authors differentiated only between current smokers and nonsmokers and, thus, inappropriately combined never smokers with past smokers.In their methods, however, Sørensen et al. reported that they originally used four categories (never smoker, occasional smoker, former regular smoker, and current smoker) to code the smoking status of the study's participants, before recoding smoking status as a dichotomous yes-or-no variable.However, as Sørensen et al. described in their statistical analysis, they analyzed their models using both of the coding methods to determine whether recoding resulted in residual confounding.Because they found no residual confounding, they chose to report results only from the model with the simpler, dichotomous coding of smoking status.
Stampfer also argues that the best way to remove residual confounding by smoking is to "simply exclude current and past smokers" [1].This exclusionary approach for smoking has been previously examined in a methodological study that utilized statistical simulation, with data from 15 diverse observational studies of body weight and mortality [3].
The study concluded that eliminating smokers from the datasets prior to analysis produces results similar to those expected from the elimination of numerically similar random proportions of the datasets prior to analysis [3].Thus, the practice of excluding smokers in studies of weight loss and mortality is highly questionable.

David F. Williamson
Centers for Disease Control and Prevention Atlanta, Georgia, United States of America E-mail: drw1@cdc.govThe research article by Pao et al. [1] provides important new information addressing three patients with acquired resistance to gefi tinib or erlotinib in progressing tumors containing a secondary mutation, leading to the substitution of methionine for threonine at position 790 (T790M) in exon 20.However, all of the patients received systemic chemotherapy prior to gefi tinib or erlotinib therapy, and the original lung tissue was obtained long before epidermal growth factor receptor (EGFR) inhibitors were used.We describe a chemonaive patient with gefi tinib-sensitive lung adenocarcinoma harboring L858R.The tumor progressed and developed an additional T790M mutation after nine months of gefi tinib treatment.
A 56-year-old female who had never smoked presented with nonproductive cough for one month.Her chest radiography revealed a mass in the right lower lung (RLL) (Figure 1A).Chest tomography (CT) confi rmed a mass with pleural invasion and multiple small nodules in the bilateral lungs.Ultrasound-guided percutaneous transthoracic lung biopsy revealed adenocarcinoma.Gefi tinib (250 mg/day) was initiated.The RLL tumor decreased in size signifi cantly two months after treatment (Figure 1B).Both serum CEA and CA-199 decreased, from 4,178 ng/ml to 9.1 ng/ml and from 464 U/ml to 22 U/ml, respectively.However, the patient could not tolerate the severe side effects, including diarrhea, erythematous papules over the nasolabial areas and buttocks, and paronychia with granulation on fi ngers.Gefi tinib was withdrawn for two weeks.Then, she received gefi tinib at 250 mg on alternate days.These side effects became tolerable, and gefi tinib at 250 mg/day was resumed.Nine months after initiating gefi tinib, chest radiography revealed progression of tumor (Figure 1C).At this time, chest CT revealed tumor progression with an endobronchial tumor in the right middle bronchus.Gefi tinib was discontinued.After obtaining written, informed consent from the patient, a CT-guided lung biopsy specimen was obtained.Pathological analysis confi rmed the presence of adenocarcinoma.This patient received subsequent chemotherapy for advanced lung cancer.
Genomic DNA was extracted from the tumor specimen of an original lung biopsy and a progressive tumor biopsy specimen.The tyrosine kinase domain (exons 18-21) was amplifi ed and sequenced.Mutations were also checked against the corresponding DNA from blood lymphocytes at the diagnosis of lung cancer.The original diagnostic biopsy specimen contained a thymidine to guanine mutation at nucleotide 2573 of exon 21, resulting in L858R.In the second biopsy, an additional single-base change from cytosine to thymidine was identifi ed at nucleotide 2369 in exon 20, resulting in T790M.
This report strengthens the evidence of T790M as an acquired gefi tinib-resistant mutation.Gefi tinib responsiveness results in large part from the drug's effective inhibition of essential antiapoptotic signals transduced by the mutant receptor, and L858R is the most commonly detected mutation [2][3][4][5].The T790M mutation is rarely found in tumors from patients not treated with either gefi tinib or erlotinib, and could be discovered only in progressing tumors, in addition to a primary drug-sensitive mutation in EGFR.A non-small-cell lung cancer cell line bearing both T790M and L858R mutations was approximately 100-fold less sensitive to gefi tinib or erlotinib, and did not show inhibition of tyrosine phosphorylation in vitro [1].
Pao et al. and Kobayashi et al. identifi ed four cases with lung adenocarcinoma harboring pre-existing mutations of EGFR as delL747-E749 plus A750P, delE746-A750, or delL747-S752, prior to the use of gefi tinib or erlotinib [1,6].All of them had exposure to previous systemic chemotherapies and took a small-molecule tyrosine kinase inhibitor as the second-or third-line therapy, then all acquired a second mutation T790M in the following months after disease progression.In the case of our patient, the patient received no prior systemic chemotherapy, and we identifi ed an initial gefi tinib-sensitizing L858R EGFR mutation, followed by a T790M mutation concomitantly with L858R in the biopsy taken from the growing tumor nine months after gefi tinib use.Though it is unlikely that prior chemotherapy led to the development of T790M mutation, given the complexity of EGFR mutation, further studies are still required to elucidate the role of T790M mutation in the context of EGFR mutations.
This correspondence was peer reviewed.

Citation:
Hambidge D (2005) Secondary schizophrenia.PLoS Med 2(9): e279.Copyright: © 2005 Dave Hambidge.This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Competing Interests: The author has declared that no competing interests exist.DOI: 10.1371/journal.pmed.0020279

Citation:
Herxheimer A (2005) Tamoxifen and the singing voice.PLoS Med 2(9): e310.Copyright: © 2005 Andrew Herxheimer.This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.Competing Interests: The author has declared that no competing interests exist.DOI: 10.1371/journal.pmed.0020310Acquired Gefi tinib-Resistant Mutation of EGFR in a Chemonaive Lung Adenocarcinoma Harboring Gefi tinib-Sensitive Mutation L858R Chien-Hung Gow, Jin-Yuan Shih, Yih-Leong Chang, Chong-Jen Yu

Figure 1 .
Figure 1.Chest RadiographyChest radiography shows a large mass in the RLL before gefi tinib treatment (A), and marked decrease in tumor size two months after gefi tinib was initiated (B).This tumor progressed nine months after gefi tinib treatment (C).