J Biomed Inform. 2010 Jun;43(3):376-84. Epub 2010 May 1.

1

Predicting Kinase Inhibition/Adverse Event Relationships with in vitro

Kinome and Clinical Trial Data to Advance Clinical Development

of New Therapeutics

Xinan Yang, PhD

1,2, Yong Huang, MD, MSc

1,2, Matthew Crowson

1,2, Jianrong Li, MSc

1,2,

Michael L. Maitland, MD, PhD3,4,5,*

, Yves A. Lussier, MD1,2,4,5,6,7,*

1Center for Biomedical Informatics,

2Sections of Genetic Medicine and

3Hematology/Oncology,

Department of Medicine, 4 UC Comprehensive Cancer Center,

5Committee on Clinical Pharmacology and

Pharmacogenomics, 6Institute for Genomics and Systems Biology,

7Computation Institute,

The University of Chicago, Chicago, IL; *Corresponding Authors

* Corresponding authors:

Yves A. Lussier, M.D.

Address: 5841 S. Maryland Avenue, MC 6092, Chicago, IL 60637-1470, USA

Tel.: (+1)773-834-0743; Fax: (+1)773-702-2567

E-mail address: ylussier@medicine.bsd.uchicago.edu

Michael L Maitland, M.D.

Address: 5841 S. Maryland Avenue, MC 2115, Chicago, IL 60637-1470, USA

Tel.: (+1)773-702-6149; Fax: (+1)773-834-0188

E-mail address: mmaitlan@medicine.bsd.uchicago.edu

2

Abstract

Background

Kinase inhibition is an increasingly popular strategy for pharmacotherapy of human diseases. Although

many of these agents have been described as “targeted therapy”, they will typically inhibit multiple

kinases with varying potency. Pre-clinical model testing has not predicted the numerous significant

toxicities identified during clinical development. The purpose of this study was to develop a

bioinformatics-based method to predict specific adverse events (AEs) in humans associated with the

inhibition of particular kinase targets (KTs).

Methods

The AE frequencies of protein kinase inhibitors (PKIs) were curated from three sources (PubMed,

Thompson Physician Desk Reference and PharmGKB), and affinities of 38 PKIs for 317 kinases,

representing > 50% of the predicted human kinome, were collected from published in vitro assay results.

A novel quantitative computational method was developed to predict associations between KTs and AEs

that included a whole panel of 71 AEs and 20 PKIs targeting 266 distinct kinases with Kd < 10uM. The

method calculated an unbiased, kinome-wide association score via linear algebra on (i) the normalized

frequencies of AEs associated with 20 PKIs and (ii) the negative log-transformed dissociation constant of

kinases targeted by these PKIs. Finally, a reference standard was calculated by applying Fisher’s exact

test to the co-occurrence of indexed Pubmed terms (p≤0.05, and manually verified) for AE and associated

kinase targets (AE-KT) pairs from standard literature search techniques. We also evaluated the

enrichment of predictions between the quantitative method and the literature search by Fisher’s Exact

testing.

Results

We identified significant associations among already empirically well established pairs of AEs (e.g.

diarrhea and rash) and KTs (e.g. EGFR). The following less well recognized AE-KT pairs had similar

association scores: diarrhea-(DDR1;ERBB4), rash-ERBB4, and fatigue-(CSF1R;KIT). With no filtering,

the association score identified 41 prioritized associations involving 7 AEs and 19 KTs. Among them, 8

associations were reported in the literature review. There were only 78 out of a total of 4,522 AE-KT

pairs meeting the evaluation threshold, indicating a strong association between the predicted and the text

mined AE-KT pairs (p= 3x10-7

). As many of these drugs remain in development, a larger volume of more

detailed data on AE-PKI associations is accessible only through non-public databases. These prediction

models will be refined with these data and validated through dedicated prospective human studies.

Conclusion and future directions

Our in silico method can predict associations between kinase targets and AE frequencies in human

patients. Refining this method should lead to improved clinical development of protein kinase inhibitors,

a large new class of therapeutics.

Keywords:

Adverse event; toxicity; kinome; kinase inhibitor; computational modeling; translational bioinformatics

Abbreviation:

PKI: Protein Kinase Inhibitor

KT: Kinase Target

AE: adverse event

Kd: dissociation constant

3

1. INTRODUCTION Motivation: Protein kinase inhibitors (PKIs) are a new class of drugs. Directly or indirectly, kinases

regulate nearly every process of cells and tissues. Scientists first described the structure and function of

protein kinases through their studies of the molecular basis of cancer. Many of the first discovered

“oncogenes” proved to be mutant forms of the genes encoding protein kinases that resulted in

autonomous or dysregulated kinase activity. Based on these concepts, PKIs were initially developed to

treat cancer, and the first such drug to be approved for marketing by the U.S. Food and Drug

Administration was imatinib (Gleevec™) for treatment of chronic myelogenous leukemia (CML).

Touted as “targeted therapy”, imatinib revolutionized treatment and prognosis for this disease[1]. But

characteristic of PKIs, imatinib not only inhibits the Bcr-Abl oncogene specific to CML, but also other

kinase targets (KTs) such as c-Kit, and platelet derived growth factor receptor alpha (PDGFRA). As

imatinib was used in increasing numbers of patients, unexpected adverse events (AEs) such as

hypophosphatemia[2] and cardiac dysfunction[3] were identified. This pattern of inhibition of both

intended and unintended KTs with clear therapeutic benefit limited by unexpected AEs has been

reproduced with every PKI approved for widespread use. Many similar agents have failed or been stalled

during clinical development because of insufficient benefit and unexpected toxicities. Better

understanding of the physiologic consequences of inhibiting specific KTs should lead to development of

safer cancer therapeutics, more effective combination treatments, and support the cross-purposing of these

drugs from cancer care for use in other diseases[4]. Specifically, confirming the absence of an adverse

event is challenging in conventional pre-clinical trials because these trials may not be conducted in animal

species exhibiting the AE. For example, rodents may be used to defined efficacy; however, this specie is

highly resistant to nausea and exotic animal species are required to determine AEs that could occur in

human. Consequently, the development of a high throughput computational method over in vitro assays

may offer the opportunity to specific the requirements of specific pre-clinical assays to rule out putative

AE identified in silico, accelerating the drug development process and reducing costs by avoiding

discovery of unexpected AEs of a compound in clinical trials.

Biological Background: Protein kinases constitute much of the signaling pathways and interactive

cellular signaling networks. More than 500 protein kinases are encoded in the human genome. The

collection of these evolutionarily conserved, modularly structured enzymes is referred to as the

kinome[5]. Given the evolutionary history of these proteins reflected in shared sequences, structures, and

functions, it is not surprising that drugs screened for capacity to inhibit specific kinases inevitably seem

also to inhibit some other kinases unintentionally. Specificity has typically been assessed in in vitro

cellular assays. These assays are typically not standardized among laboratories and cross-reacting kinases

might not be relevant to cellular function or even be expressed at all in the particular tested cells.

Furthermore, the non-conserved sequences of kinase genes contribute to the expression and functional

differences in cellular signaling that produce the metabolic and physiologic differences among species, so

animal toxicity findings do not consistently predict effects in humans. Consequently, the introduction of

PKIs to human subjects has led to novel observations not only for clinical therapeutics but for better

understanding human molecular physiology[4].

Although many methods have been developed[6, 7] to predict chemical structure/kinase inhibition

relationships, there has been no dedicated effort to associate inhibition of specific kinases with

physiologic consequences through systems analysis approaches. Two major obstacles have been

observed 1) the limited, labor-intensive assessment of PKI specificity through cellular assays without

central laboratory standardization, and 2) limited concurrent comparisons of PKIs in clinical use or

development on the same platform. Recently, Zarrinkar and colleagues presented the largest comparative

analysis of PKI selectivity (including 38 PKIs) with an unbiased in vitro kinome-wide binding assay[8].

Although with acknowledged limitations, their data provide the opportunity to cross-compare effects of

4

many PKIs in clinical use and development. The structured reporting of AEs in the clinical trials that have

tested these agents provides a database with which to infer AE-KT relationships.

Table 1. The comparison with related studies

AE-

Drug

Drug Target Prediction Protein target-

Toxicity

PKI-Specific Input

Data Size

Lit

era

ture

Ph

ys

ical

Kin

om

e

Lit

era

ture

Ch

em

ical

Str

uc

ture

Co

mp

t .

Pre

dic

tio

n

Ev

alu

ati

0o

n

Kin

om

e-w

ide

An

aly

sis

Lit

era

ture

Co

mp

t. P

red

icti

on

Ev

alu

ati

on

Ad

ve

rse

Eve

nts

(#

)

Kin

as

e In

hib

ito

rs (

#)

Kin

as

e t

arg

ets

(#

)

Studies Focusing on kinanes Toxicities

Present study ■ ■ ■ ■ C,M 71 20 266

General Studies predicting target Toxicities (not focused on kinases)

Hansen 2009[9] ■ ■ B ■ 3 1* -

Bender A, et al. 2007[10] ■ ■ ■ ■ 166 - 0*

DART 2003[11] ■ ■ ■ 187 - -

Apic 2005 [12] ■ ■ - - -

Studies predicting drug targets, but not their Toxicities

Not Applicable

Fliri 2007 [13] ■ ■ C,B 5,923 2* 168*

Garten 2009 [14] ■ 395*a 21*

b 121*

c

Liebovitch 2007 [15] ■ - - -

Campillos M, et al. 2008[16] ■ ■ ■ B 727 3* -

LEGEND: C=computational evaluation, B=biological validation, M= manual curation, #: count,

*: subset of gene targets with protein kinase activity (GO:0004672) or subset of drugs that are kinase inhibitors;

- : no specific detail about the input data

a: by searching the key words “side effect” in the Pharmspresso on line database

(http://pharmspresso.stanford.edu/ygarten/Pharmspresso/html/index.html)

b: by searching the key words “human”, “kinase” and “Inhibitor” in the Pharmspresso on line database

c: by searching the key words “human” and “kinase” in the Pharmspresso on line database

Bioinformatics Background: Although there exist many computational approaches to predict individual

drug targets, few studies pertain to many targets. We summarized in Table 1 those studies that access

multiple-target predictions (high throughput). Some excellent studies focused on computationally

predicting the target of drugs (bottom part of the table), while others focused on their toxicities (middle

part). On the first row, the present study, focuses on kinases, uses the physical kinome map and provides

an evaluation of adverse event relationships, elements that contrast with the majority of other studies.

Our methodological approach also differs from that of these other publications. First, we used text

mining technology to evaluate our computationally predicted results. A few others undertaking such

5

large-scale approaches have incorporated text-mining and natural language processors to unlock relevant

data from disparate and heterogeneous sources. For example, Kuhn et al. have created a search tool for

interactions of chemicals and proteins, coined “STITCH”, that consolidates chemicals and draws

relationships between the chemicals and their activity data in cell lines, MeSH assignments, and

literature[17]. Textpresso and other text mining approaches were used to curate the relationships between

drugs and genes[11, 14]. However, simple text-mining would not provide enough statistical power for a

group of new drugs under active clinical development, since there is insufficient openly accessible

literature. For example, searching Pharmspresso by “EGFR side effect” results in no match[14]. Second,

we are the first to perform prediction analyses based on kinome-wide physical mapping. Prior studies of

adverse effects have mostly generated or used molecular pathway atlases to assess structure-function

correlations in complex systems. For example, by understanding the physiological pathways and

potential binding partners of a candidate drug, it is theoretically possible to anticipate adverse events of

the candidate if the consequences of binding to these non-target partners are known. One commercial

provider, Cambridge Cell Networks (CCNet) has created the “PathTox” tool that offers a sequence search

in their pathway atlas that reports the probable secondary effects of candidates binding to specific

partners[12]. Further, a research group at Pfizer has been able to forecast the effects of drugs on a large

scale using computational techniques to establish relationships between structure and function to predict

the probability of two drugs exhibiting similar system-wide effects[13]. A recent study based on

integration of data on gene-drug interactions, gene-interaction and drug-drug similarity predicted novel

candidate genes that might affect interindividual differences in metabolism, effectiveness, or adverse

events for four drugs including one PKI gefitinib[9]. Third, though the feasibility of using adverse

events to reveal the molecular and genetic interactions with drugs in humans has already been established,

we are the first to develop computational methods focused on the complex, often overlapping

binding/inhibition profiles of PKIs specifically. A relevant recent study uncovered 12 drug-target

relationships by means of predicting relationships using phenotypic adverse event similarities; they

further validated all of these relationships and successfully confirmed 9 of the relationships using in vitro

binding assays[16]. This study is the first to predict computationally binding of drugs to unintended target

molecules based on correlations between target binding and adverse event similarity. Others predicted

combination of drugs with desired effects based on chemical structures of drugs[15]. A pilot study

predicted toxicity related targets by combining drug off-targets binding and adverse reactions using

Bayesian methods[10], however, their resulting 70 targets contain no kinase targets.

Rationale: To predict computationally the toxicities resulting from inhibition of specific KTs we

developed a novel quantitative method which was designed to be comprehensive and unbiased, including

all available information on AEs and PKIs. It provided a proof-of-concept that AE-KT relationships can

be predicted by analyzing the specific AEs induced by multiple inhibitors and the propensity of these

inhibitors to bind specific KTs on a physical kinome map. We then evaluated the prediction method by

comparing the results to evidence from literature mining.

Contribution: To our knowledge, we have performed the first computational prediction of toxicity of

kinase inhibition using kinome-wide physical mapping. Our work contributed a table of PKIs with

reported adverse events and their prevalence.

2. METHODS In this paper, we aim to offer a proof-of-concept by presenting a significant correlation between

frequencies of adverse events and kinome inhibition patterns through a survey of emerging kinase

inhibitors in 3 modules: 1) Literature mining to generate quantity mapping between PKIs and adverse

events; 2) Computationally predicting the significant association between targets and adverse events using

two different methods; 3) Evaluating the predicted results by independent literature searching of co-

occurrence and significance estimation. Figure 1 introduces the data and their relationship and Figure 2

demonstrates the three modules.

6

Figure 1. Illustration of raw data and their relationship in this study. The frequencies of adverse events (AEs)

associated with kinase inhibitors were curated from three sources (Thompson Physician Desk Reference, PharmGKB and PubMed). The dissociation constants (Kd) of kinases were summarized from physico-chemical assays[8]. The

computational method in this study considers all of these data to predict the association between AEs and targets of kinase inhibitors.

2.1 Data Organization 2.1.1 Kinome Map Collection and Procession

The physical kinome map was summarized from novel PKI-KT dissociation constant (Kd) data for 38

kinase inhibitors against a panel of 317 kinases[8]. Three steps were performed before statistic

computation: First, mutated targets were excluded, except for those without assay for their non-mutation

targets which were GCN2, JAK1, JAK2 and JAK3, since the homologous gene targets (mutated kinase)

are repeated measures of the related targets comparing to those of more independent genes. Second, 20

PKIs having no curated AEs were excluded. Finally, we took a negative log-transformation of all Kd

values, because logarithm transformation reduces detection noise to an additive level and tends to convert

exponential distribution trends to normal distribution trends (Suppl. Figure 1).

Figure 2 The study consists of three modules: Data organization, computational prediction and computational evaluation. In the module of Data organization, two steps generated two different associations used

as the inputs to the “prediction of kinase toxicity” module. In step , a dissociation constant is found in a physical kinome map, while in step , drug adverse events were curated from multiple sources. In the modules of prediction of

7

kinase toxicity ( step ), a quantitative method was developed to identify the prioritized AE-KT association scores (PAS). In the evaluation ( step ), Pubmed was mined to identify enriched AE-KT pairs from a total of 4,522 putative pairs among 17 AEs that have more than five PKI repeats and 266 observed kinases. Then the findings of the predictive method were evaluated against the text mining results.

2.1.2 Curation for Relationship between PKIs and Adverse Events from various Knowledge

Sources (Suppl. Table 1)

Adverse Events Source Selection: To locate the adverse events data for the collection of the 38 kinase

inhibitors[8] with kinase binding information, we reviewed a clinical pharmaceutical resource - Thomson

Physician’s Desk Reference (PDR), a biomedical informatics resource - PharmGKB, as well as Medline-

indexed primary research articles via PubMed. Given the heterogeneity of the PKI data reported in our

sources, data found in PDR and PharmGKB was given priority over any journal-derived clinical data.

When we failed to find an agent in either PDR or PharmGKB, the most recent journal articles with the

most explicitly quantified adverse event data were used. To cross-check the validity of the collected

adverse event frequency values in PDR and PharmGKB, we also compared the adverse event frequency

data for PKIs found in the electronic PKI database Facts & Comparisons (Version 4.0) for consistency.

Adverse Event Normalization: We normalized both the frequency values and the symptom/adverse

event terms of the adverse event data extracted from PharmGKB, PDR, and recent literature. We reported

AE frequencies as percentage values. First, to normalize AE frequencies reported in the literature as a

range or as different frequencies from various sources, we took the mean value of the range. And for

frequencies presented in a series of comparative doses (e.g. 10mg vs. 100mg vs. 200mg) we used the AE

frequencies associated with the highest dose. Second, to normalize the AE terms we created logical AE

terms to consolidate related AEs (e.g. “Rash” encompasses “skin Rash”, "vesiculobullous rash",

"acneiform rash", and “maculopapular rash”). For simplification, we hereafter refer to these normalized

AE as AE.

Expected Adverse Event Frequency in Population: In practice, a universal threshold indicating a

frequency above that of placebo for all adverse events was inappropriate. Indeed, certain adverse events

like headache or back pain occur at much higher incidence in the population than that of neutropenia.

Therefore, we reviewed the literature to find out the expected frequency, and then decided a little severer

threshold (generally by 50-100% above expected frequency) for every adverse event. The literature

sources comprise of: 1) wrong diagnosis (http://www.wrongdiagnosis.com/p/pain/prevalence-types.htm);

2) emedicine.com (http://www.emedicine.com/emerg/topic233.htm); 3) population health metrics

(http://www.pophealthmetrics.com); 4) PKIs.com (http://ww.PKIs.com) 5) DailyMed

(http://dailymed.nlm.nih.gov); 6) Merck manual online; 7) google “adverse event” prevalence.

2.2 Computational Prediction on Associations Between Adverse Events and the

Kinome The method was designed to take into account every piece of evidence from the adverse events, kinases

and kinome map. The inputs were two quantitative association matrices: the curated AE-PKI frequency

matrix and the PKI-KT binding affinity matrix, respectively (Figure 2).

We introduce definitions as follows: (dx), x=1,…,p, denotes a set of protein kinase inhibitor s(PKIs); (ty ),

y=1,…,q, denotes a set of kinase targets (KTs); and (ez), z=1,. . .,r, is a set of adverse events (AEs).

Definition 1. The AE-KT association matrix A=(azy) consists of an association score that reflects the

association between the kinase target ty and the adverse event ez which is a (r × q) dimension matrix.

Definition 2. The AE-PKI association matrix F=(fzx ), a (r × p) dimension matrix, consists of the

frequency of normalized adverse event ez for a PKI dx which were curated from three sources (PubMed,

Thompson Physician Desk Reference and PharmGKB).

8

Definition 3. The PKI-KT relationship matrix K=(kxy) , which is a (p × q) dimension matrix, consists of

an experimental dissociation constant value (Kd value) for a PKI dx targeting ty that was reported in

literature[8].

Prioritized Association Scores (PAS)

The PAS method assumes that the toxicity of a single kinase can be repeatedly observed by multiple PKIs

targeting this kinase with respects to both their combining affinity and their AE frequencies. By an

association operation, the matrix multiplication, we scored the AE-KT associations (Figure 3) as:

p

x xyzx kfKFA1

)log()()log( (1)

where F and K were the normalized frequency of AEs and the dissociation constant of PKIs, respectively.

The higher the score, the more prioritized the association of the kinase with the corresponding adverse

event would be. Following the study of the false discovery rate, a discovery threshold was set as the top

(0.2% quantile) of all observed scores in this study. Moreover, we traced back to identify the PKIs that

associated both to the kinase and the adverse event of finding. These PKIs are much of interests because

they should be avoided of together using due to the higher possibility of inducing the same adverse event.

False discovery rates were computed on each score by permutation resampling of K. Multiple

comparisons which were adjusted using the q-values[18, 19]. R language source of the Prioritized

Association Scores method is available at http://www.lussierlab.org/publication/PAS/ .

Figure 3. The association scores identified prioritized AE-KT pairs were included to generate the output bipartite network. The AE-KT prioritized association scores in matrix A were constructed by applying a matrix multiplication on the AE-PKI frequency matrix F and the PKI-KT binding affinity matrix K. A prioritized AE-KT pair has an association score higher than the top (0.2% quantile) of all observed scores (q-value < 21%).

2.3 Evaluation Text Mining of PubMed: To predict the adverse event of kinase by significant co-occurrence in abstracts

indexed in PubMed, we used Pubmatrix[20], a multiplex literature mining tool. Let N be the totally

records in PubMed, we created a contingency table (Table 2) for each putative pair of AE-KT: n2 is the

number of abstracts containing a kinase t, n3 is the number of abstracts about an adverse event e, and n1

is the number of abstracts of both. Then one-sided Fisher’s exact test was applied to evaluate the

significance of association between each pair of kinase target and adverse event[7, 21]; and a gold

standard of unadjusted p ≤ 0.05 was set as significant for a co-occurrence. We further manually verified

the computationally corroborated pairs and rejected targets whose symbols appearing in abstracts refer to

other objects instead of genes, e.g. CIT is the abbreviation for “city”, YES refers to “yes”, MET refers to

“Methods” or “metabolic equivalents”, Gak as institute name, and TEC is the abbreviation for “toxigenic

Escherichia coli” or “total eosinophil counts” in many abstracts.

Table 2. Contingency table for each AE-KT pair using text mining of Pubmed

#ref. include kinase t #ref. not include kinase t ∑

9

#ref. include AE e n1 n3-n1 n3

#ref. not include AE e n2-n1 N-n2-n3+n1 N-n3

∑ n2 N-n2 N

Using a Fisher’s Exact Test, we further evaluated the enrichment (overlap) between the AE-and-KT co-

occurrences in pubmed identified by computational text mining and the AE-KT pairs predicted by the

association score method conducted over the kinome map.

3. RESULTS

3.1 Mapping PKIs and adverse events using literature curation Karaman MA et al. successfully mapped an interaction for 38 PKIs (drugs) across a panel of

317 kinases[8]. Our literature curation resulted in 71 normalized AEs induced by 20 of these 38 PKIs

from 181 literature-reported adverse drug activities (Figure 2). The primary analysis was performed on

the minimal panel of the 266 distinct kinases targeted by 20 PKIs, whereas the 20 PKIs had 71 reported

adverse events. Suppl. Table 1 gives the details of normalized adverse events, their prevalence and the

estimated threshold of frequency. In the table, the adverse event “Asthenia” is merged with “Fatigue”

because of the similarity of these two categories; conversely, the adverse event “Pain” is further

subdivided into “Back Pain”, “Chest Pain” and “Pain” because of their different ratios of incidences. A

smaller panel of inputs for evaluation, the 17 AEs with more than 5 repeats of PKIs that target 266

distinct kinases is described in Suppl. Table 2.

3.2 Computational prediction of significant associations between AEs and kinase

targets (Methods, Formula 1). . We identified 41 prioritized AE-KT associations among 19 targets and 7 adverse events using a

threshold of the top of 0.2% quantile of all 18,866 (71x266) scores (Suppl. Figure 2 and Suppl. Table

3). Eight prioritized associations by PAS were also independently related to significant co-occurrences

identified by text mining: nausea to ERBB2, fatigue to VEGFR2, diarrhea to both KIT and EGFR, rash to

both EGFR and ERBB2, and asthenia to both ERGF and ERBB2 (red lines in Suppl. Figure 2 and green

lines in Suppl. Figure 3). Figure 4 is a bipartite network of the identified associations summarized from

the associations between KTs and AEs after merging the vertices of similar AEs into one vertex, which

are nausea and vomiting, asthenia and fatigue.

Kinase

AE

Line Width : Prediction Score (a)

a > 0.2 (median q < 1%)

0.2 ≥ a > 0.113 (q ≤ 21%)

Line Color: Evaluation in Review of Literature

red Also found in Literature Review

grey Not found in Literature Review

Figure 4. The predicted network of adverse events of kinase targets using association score. Bipartite network

of 41 prioritized AE-KT pairs with the top PAS association scores (a score > 0.2; the best 1/5 of 1% of all calculated

10

scores) consisting of 7 distinct AEs and 19 distinct kinase targets. As shown in the plot, a wider line indicates a higher priority of association. In additional, the predicted AE-KT associations by text mining of PubMed (Fisher’s exact test p≤0.05) are resented in red color. In our evaluation using text mining, a Fisher’s Exact test showed a significant (p=3x10

-7) association between the literature review among these 41 predicted AE-KT pairs. Pairs of similar vertices

(Nausea and Vomiting, Asthenia and Fatigue) were each merged into a single vertex, respectively (For more details, see Suppl. Figure 2).

3.3 Evaluation Independently, we searched PubMed using the on line tool Pubmatrix[20] for every putative pair of

kinase targets and adverse events. For simplification, we used a smaller panel of inputs, the 17 AEs with

more than 5 repeats of PKIs that target 266 distinct kinases. This panel evaluates the literature co-

occurrence of 4,522 (266x17) putative AE-KT pairs, which excludes only 3 captured AE-KT pairs from a

total of 5,320 (266x20) putative pairs in PAS. As of Oct 21, 2008, there were a total of 19,028,626 (N)

abstracts indexed in PubMed. Among them, only 105 AE-KT pairs meet the text mining gold standard

(the unadjusted Fisher’s exact p-value ≤ 0.05). After manual verification, 27 pairs involving five targets,

which were CIT, YES, MET, Gak and TEC, were rejected due to their searched symbol representing other

objects instead of gene in the abstracts indexed in PubMed (For details, see Suppl. Table 4). Thereafter,

78 predicted AE-KT pairs were presented in as the predicted network of side effects and kinase targets

using computational text mining of literature indexed in PubMed (Suppl. Figure 3).

Eight out of 41 AE-KT pairs predicted using association score were also predicted by literature review. A

Fisher’s Exact test showed a strongly significant association between the discovery of our quantitatively

computational method and the literature review: p-value=3x10-7

.

4. DISCUSSION Our study recapitulated well recognized associations between AEs and widely used PKIs. The two

strongest relationships captured are diarrhea, a recognized, mechanism-based adverse event of epidermal

growth factor receptor (EGFR) inhibition, and the other common EGFR inhibitor-induced AE, rash. The

rash caused by EGFR kinase inhibitors has been associated with clinical response for multiple cancers,

and this serves as evidence that intended KTs have both salutary and adverse effects where the

relationship was confirmed by the PAS method[22-24]. Examination of these strongest relationships is

useful for assessing our methods development, making modifications for future analyses, and generating

hypotheses that might be easily validated on existing clinical specimens or data.

The cumulative adverse event data in this study are collected from phase I, phase II, phase III, and other

clinical investigations of numerous kinase inhibitors, both those FDA approved and those not yet

completing clinical development; a larger sample than any individual trial. These studies describe the full

spectrum of adverse events both drug-related and disease-related. In any single trial (especially if this trial

is not placebo controlled- the typical case for oncology phase I trials) it is frequently not apparent which

toxicities are reproducibly attributable to the drug, and it is certainly not clear if the adverse events are

due to the mechanism of action of the drug or an “off-target” effect. The comparison of drugs with

overlapping and non-overlapping molecular binding properties provides a rational approach to

determining which adverse events, dose-limiting or not, might be attributed to the inhibition of which

targets. Indeed, many of the so-called targeted kinase inhibitors have been shown to target a wider

spectrum of kinases than originally planned[25].

An unexpected and exciting finding is the identification of the CSF1R/fatigue and KIT/fatigue

associations. As depicted on the kinome dendrogram[26], CSF1R, KIT and VEGFR2 are three kinases

with similar sequence and physical structure but with biological roles important to different cell subsets.

VEGFR2 is almost exclusively expressed on endothelial cells and is activated in tumor angiogenesis.

Therefore this has been the primary target for several new cancer therapeutics such as sunitinib and

AMG-706 (now known as motesanib). CSF1R plays a major role in granulocyte/monocyte/macrophage

development, signaling, and regulation of cytokine release and response. KIT signaling has been reported

11

to mediate various roles in different tissues, but is most prominently involved in proliferation and

differentiation of various hematopoietic lineages and in mature mast cell and eosinophil signaling. The

agents developed to inhibit the structurally similar VEGFR2 tend to have relatively high potency for

CSF1R and KIT and fatigue is a common, sometimes dose-limiting adverse effect of these drugs.

Although we have identified a strong association between these kinases and fatigue, it remains unclear

whether one, two, or all three of these KTs mediates the fatigue. If any of the three is differentially

associated, there will be the opportunity in either development of new drugs or in administration of

combinations of PKIs to decrease the associated fatigue by either selectively eliminating the offending

KT affinity or by selecting 2 PKIs that when added together will maximize inhibition of the intended

therapeutic KT, while reducing the inhibition of the unintended, fatigue-inducing target. Given the

unbiased identification of this AE-KT association and its strength relative to the panel of relationships

tested, our clinical program has commenced studies of human circulating protein biomarkers for

disruption of CSF1R signaling and validation studies of the relationship between inhibition of CSF1R and

fatigue in cancer patients treated with PKI’s that target VEGFR2.

Mined facts from Pubmed used in the evaluation are specific for each drug and, to our knowledge, only in

very few specific cases have been generalized to a group of drugs. While this is a limitation of the

evaluation, we believe this may be the first high throughput evaluation of PKI toxicities. While the gold

standard of drug target prediction, an older field, is the biological validation in vitro, drug toxicities

predictions require a much more complex design for two reasons: in vitro cellular studies cannot reflect

more than a cellular toxicity (animal models would be required for most clinical toxicities such as rash,

vomiting, etc), and because of interspecies differences in kinase structures and functions animals may

frequently not manifest the toxicities seen in humans. Finally, a specific drug target prediction is not

generalized to a class of drug, while an adverse event associated to a kinase (target) is a generalized

“systems” property and requires a more comprehensive design for validation.

Future studies and limitation. Adverse events occur in clinical studies where the tissue perfusion of

kinase inhibitor depends on its specific pharmacokinetics, while the in vitro binding affinity of a kinase

receptor to a kinase is measured as dissociation constant Kd. Our approach provides calculations of AE

and Kd jointly to create a compound score (PAS) and control for multiplicity. Thus significant PAS

scores are associated to a specific kinase, and may imply different underlying Kd values for this kinase

under different drug AE. It is a known fact that each drug may have a specific pharmacokinetic profile,

and thus the corresponding relevant in vitro Kd remains data driven in our study. This kind of limitation

in our study is also shared with previous published efforts to use incidence of adverse events as a means

to infer previously unrecognized and untested drug/target relationships[9-11, 13, 14, 16]. An important

advance upon the current method will be to integrate data on physiologically relevant pharmacokinetic

and pharmacodynamic properties of different drugs and targets. This will be most readily accomplished

by first performing clinical validation of some of the associations identified in this study.

We plan to conduct future studies to validate and extend the method with toxicity data from clinical trials

of novel kinase inhibitors sponsored by the National Cancer Institute. Better quantitative gold standard

data for adverse events might have been obtained from the US adverse event report system (AERS) and

the Canadian adverse event reporting and learning system (CAERLS). And a standardized ontology for

adverse event is needed which might be taken from Health Canada’s Canadian Adverse Drug Reaction

Monitoring Program (CADRMP, http://www.hc-sc.gc.ca/dhp-mps/medeff/databasdon/index-eng.php).

Biological validation of any specific target cannot be conducted using this method and would be a

complex pursuit. However, for strong candidates with multiple independent, but consistent data sources,

we would consider a human clinical trial the best, gold standard approach to biological validation.

Additionally, we will explore additional models involving pathways, multi-kinase effects, and other

models of drug/kinase inhibition beyond log-linear ones. However, by design, the physical kinome map

limits the breadth of our predictions to kinase toxicities. Arguably, other non-kinasetargets may exist and

are beyond our calculations; however it will be challenging to distinguish non-target specific effects from

12

those of collective kinase inhibition. There is increasing evidence that many PKIs aren’t as specific as the

concept of “targeted therapy” implies[25].

5. CONCLUSION Predicting toxicities of kinase inhibitors is an important problem for clinical development of new drugs.

Current methods for predicting toxicities mainly rely on pre-clinical cell and animal studies[27]. Our in

silico method can predict associations between kinase targets and AE frequencies in human patients. This

method was controlled by false discovery, literature and expert review. Expert review further confirmed

that the method recapitulates existing empirical and mechanistic knowledge and provides novel and

clinically relevant insights (e.g. CSF1R/fatigue and KIT/fatigue). To our knowledge, this is the first study

using kinome binding patterns for predictions that proposes a novel Prioritized Association Score and

permutation resampling for increased sensitivity of results in spite of the scarcity of information about

adverse effects of kinase inhibitors.

Refining the method should lead to improved clinical development of protein kinase inhibitors, a large

new class of therapeutics[4]. 1) As some AE-KT relationships will only be apparent at the system

physiology level, comprehensive screening for AE-KT relationships in patients exposed to these drugs

provides the most readily available method for detecting these relationships. 2) This approach bypasses

the species differences in drug pharmacokinetics/pharmacodynamics that impair translation from animal

models. 3) Knowing AE-KT relationships can improve the safety and speed the completion of early trials

of new kinase inhibitors. For example if inhibition of a particular kinase is associated with

hypertriglyceridemia, then determining the active dose range of a novel PKI that blocks that KT could be

accelerated by measuring serum triglyceride elevations whether or not that KT is the intended target.

4) This approach also can inform selection from among the thousands of possible combinations of

currently available PKI’s, those that have the greatest likelihood for non-overlapping toxicity and hence a

better therapeutic index. 5) Finally, uncovering AE-KT relationships with novel PKI’s can reveal new

insight into the molecular physiology of the human body, leading to the identification of risk factors for

bad outcomes for some drugs and new clinical indications for others [4].

ACKNOWLEDGEMENTS This work was supported in part by the NIH/NLM/NCI National Center for Multiscale Analyses of

Genomic and Cellular Networks (MAGNET, 1U54CA121852), the NIH/NCRR Clinical & Translational

Science Awards (1U54 RR023560-01A1), The Cancer Research Foundation, The University of Chicago

Cancer Research Center, and The Ludwig Center for Metastasis Research. MLM is supported by

Mentored Career Development award K23CA124802. XY is partly supported by the Natural Science

Foundation of China 60971099.

REFERENCES [1] Druker BJ, Guilhot F, O'Brien SG, Gathmann I, Kantarjian H, Gattermann N, Deininger MW, Silver

RT, Goldman JM, Stone RM, Cervantes F, Hochhaus A, Powell BL, Gabrilove JL, Rousselot P,

Reiffers J, Cornelissen JJ, Hughes T, Agis H, Fischer T, Verhoef G, Shepherd J, Saglio G, Gratwohl

A, Nielsen JL, Radich JP, Simonsson B, Taylor K, Baccarani M, So C, Letvak L, Larson RA. Five-

year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 2006;355:

2408-17.

[2] Berman E, Nicolaides M, Maki RG, Fleisher M, Chanel S, Scheu K, Wilson BA, Heller G, Sauter NP.

Altered bone and mineral metabolism in patients receiving imatinib mesylate. N Engl J Med

2006;354: 2006-13.

13

[3] Kerkela R, Grazette L, Yacobi R, Iliescu C, Patten R, Beahm C, Walters B, Shevtsov S, Pesant S,

Clubb FJ, Rosenzweig A, Salomon RN, Van Etten RA, Alroy J, Durand JB, Force T. Cardiotoxicity

of the cancer therapeutic agent imatinib mesylate. Nat Med 2006;12: 908-16.

[4] Maitland ML, Ratain MJ. Terminal ballistics of kinase inhibitors: there are no magic bullets. Ann

Intern Med 2006;145: 702-3.

[5] Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the

human genome. Science 2002;298: 1912-34.

[6] Muster W, Breidenbach A, Fischer H, Kirchner S, Muller L, Pahler A. Computational toxicology in

drug development. Drug Discov Today 2008;13: 303-10.

[7] Agarwal P, Searls DB. Literature mining in support of drug discovery. Brief Bioinform 2008.

[8] Karaman MW, Herrgard S, Treiber DK, Gallant P, Atteridge CE, Campbell BT, Chan KW, Ciceri P,

Davis MI, Edeen PT, Faraoni R, Floyd M, Hunt JP, Lockhart DJ, Milanov ZV, Morrison MJ, Pallares

G, Patel HK, Pritchard S, Wodicka LM, Zarrinkar PP. A quantitative analysis of kinase inhibitor

selectivity. Nat Biotechnol 2008;26: 127-32.

[9] Hansen NT, Brunak S, Altman RB. Generating genome-scale candidate gene lists for

pharmacogenomics. Clin Pharmacol Ther 2009;86: 183-9.

[10]Bender A, Scheiber J, Glick M, Davies JW, Azzaoui K, Hamon J, Urban L, Whitebread S, Jenkins JL.

Analysis of pharmacology data and the prediction of adverse drug reactions and off-target effects

from chemical structure. ChemMedChem 2007;2: 861-73.

[11] Ji ZL, Han LY, Yap CW, Sun LZ, Chen X, Chen YZ. Drug Adverse Reaction Target Database

(DART) : proteins related to adverse drug reactions. Drug Saf 2003;26: 685-90.

[12] Apic G, Ignjatovic T, Boyer S, Russell RB. Illuminating drug discovery with biological pathways.

FEBS Lett 2005;579: 1872-7.

[13] Fliri AF, Loging WT, Volkmann RA. Analysis of system structure-function relationships.

ChemMedChem 2007;2: 1774-82.

[14] Garten Y, Altman RB. Pharmspresso: a text mining tool for extraction of pharmacogenomic concepts

and relationships from full text. BMC Bioinformatics 2009;10 Suppl 2: S6.

[15] Liebovitch LS, Tsinoremas N, Pandya A. Developing combinatorial multi-component therapies

(CMCT) of drugs that are more specific and have fewer side effects than traditional one drug

therapies. Nonlinear Biomed Phys 2007;1: 11.

[16] Campillos M, Kuhn M, Gavin AC, Jensen LJ, Bork P. Drug target identification using side-effect

similarity. Science 2008;321: 263-6.

[17] Kuhn M, von Mering C, Campillos M, Jensen LJ, Bork P. STITCH: interaction networks of

chemicals and proteins. Nucleic Acids Res 2008;36: D684-8.

[18] Storey JD. A direct approach to false discovery rates. Journal of the Royal Statistical Society, Series

B 2002;64: 479-498.

[19] Storey JD. The positive false discovery rate: A Bayesian interpretation and the q-value. Annals of

Statistics 2003;31: 2013-2035.

[20] Becker KG, Hosack DA, Dennis G, Jr., Lempicki RA, Bright TJ, Cheadle C, Engel J. PubMatrix: a

tool for multiplex literature mining. BMC Bioinformatics 2003;4: 61.

[21] Fisher RA. On the Interpretation of χ2 from Contingency Tables, and the Calculation of P. Journal of

the Royal Statistical Society 1922;85: 87-94.

[22] Cohen EE, Halpern AB, Kasza K, Kocherginsky M, Williams R, Vokes EE. Factors associated with

clinical benefit from epidermal growth factor receptor inhibitors in recurrent and metastatic

squamous cell carcinoma of the head and neck. Oral Oncol 2009.

[23] Perez-Soler R, Saltz L. Cutaneous adverse effects with HER1/EGFR-targeted agents: is there a silver

lining? J Clin Oncol 2005;23: 5235-46.

[24] Wacker B, Nagrani T, Weinberg J, Witt K, Clark G, Cagnoni PJ. Correlation between development

of rash and efficacy in patients treated with the epidermal growth factor receptor tyrosine kinase

inhibitor erlotinib in two large phase III studies. Clin Cancer Res 2007;13: 3913-21.

14

[25] Fojo T. Commentary: Novel therapies for cancer: why dirty might be better. Oncologist 2008;13:

277-83.

[26] Miller ML, Jensen LJ, Diella F, Jorgensen C, Tinti M, Li L, Hsiung M, Parker SA, Bordeaux J,

Sicheritz-Ponten T, Olhovsky M, Pasculescu A, Alexander J, Knapp S, Blom N, Bork P, Li S,

Cesareni G, Pawson T, Turk BE, Yaffe MB, Brunak S, Linding R. Linear motif atlas for

phosphorylation-dependent signaling. Sci Signal 2008;1: ra2.

[27] Shuler ML. Animal Surrogate Systems for Toxicity Testing In: McAuliffe GJ, Tatosian DA, editors.

Encyclopedia of Biomaterials and Biomedical Engineering 2006.

15

List of Supplementary Files

Supplementary Figure 1 - Histogram of raw Kd values (KD) and the log transformed Kd values

(KD.log).

Supplementary Figure 2 - Bipartite network of the prioritized AE-KT pairs identified using association

score.

Supplementary Figure 3. The predicted network of adverse events of kinase targets using text mining in

PubMed.

Supplementary Table 1 - The 72 normalized adverse events, their prevalences and thresholds of

frequencies.

Supplementary Table 2 - The 17 normalized adverse events for 20 PKIs and their frequencies (F) in

percentage.

Supplementary Table 3 - The 41 prioritized AE-KT pairs identified using prioritized association score.

Supplementary Table 4 - The manually verification resulted in 78 out of 105 significant (unadjusted p-

value≤0.05) AE-KT pairs that were identified using text mining in PubMed.

J Biomed Inform. 2010 Jun;43(3):376-84. Epub 2010 May 1.

16

Supplementary Figures

Supplementary Figure 1 - Histogram of raw Kd values (KD) and the log transformed Kd values (KD.log). It

shows that logarithm transformation tends to convert the values to be analyzed from an exponential distribution to a

normal distribution.

17

Kinase

AE

Line Width : Prediction Score (a)

a > 0.2 (median q < 1%)

0.2 ≥ a > 0.113 (q ≤ 21%)

Line Color: Evaluation in Review of Literature

red Also found in Literature Review

grey Not found in Literature Review

Supplementary Figure 2 – The predicted bipartite network of the prioritized AE-KT pairs using association

score. This network consists of 7 distinct AEs and 19 distinct kinase targets. As shown in the plot, a wider line

indicates a higher priority of association. The corresponding PAS prediction score (a) is labeled aside each AE-KT

linkage. Additionally, the AE-KT associations corroborated by literature review (Fisher’s exact test p≤0.05, and

manually verified) are presented in red. A Fisher’s Exact test showed a strongly significant (p=3x10-7

) association

between the discovery of the asssociation score and the literature review among these 45 predicted AE-KT pairs.The

maximum observed PAS score in our study was 0.33.

18

Supplementary Figure 3. The predicted network of adverse events of kinase targets using text mining in

PubMed. The resulting 78 significant co-occurrences (p≤0.05, unadjusted Fisher’s Exact test on adverse events per

kinase; see Methods) consist of 14 AEs (appear in triangle) and 40 kinases (appear in circle). As shown in the plot,

the PAS predicted AE-KT pairs are presented in green. Since there are eight out of 41 predicted AE-KT pairs

recognized by text mining, a Fisher’s Exact Test showed a strongly significant over-representation between the

discovery of our quantitative method and the literature review for all 4,522 AE-KT pairs (p-value=3x10-7; see

Methods). Pairs of similar vertices (Nausea and Vomiting, Asthenia and Fatigue) were each merged into single

vertex, respectively.

J Biomed Inform. 2010 Jun;43(3):376-84. Epub 2010 May 1.

19

Supplementary Table 1 - The 72* normalized adverse events, their prevalences and thresholds of frequencies.

Normalized Adverse

event Other Terms Included Prevalence Reference

Abdominal Distention 1.65% USA [1]

Abdominal Pain "abdominal pain or cramping";

"Pain, abdomen"

Mexico (46%) Brazil (43%) Japan (10%); 12.5%

placebo ; 0 placebo

Drug.com[2], [3],

[4]

Acne 6.25% USA [5]

Alopecia 0.15% in UK, incidence is 0.05-2% [6]

Altered taste "taste disturbance" ;"dysgeusia" <3%

Amblyopia >3%; 11.4% low income school children, Brazil. [7, 8]

Anemia 1.29% USA [9]

Anorexia 0.10% USA; 1.6% placebo [10]; Drugs.com[2]

Arrhythmia 5.3% USA; 0 placebo [11]

Arthralgia "joint pain"; "Pain, joint" >7 million USA; 6.8% woman 16.7% health pregnant

women [12]

Ascites

Asthma 6.4% USA;5.1% NHIS95;5 population (NWHIC) [13]

Atrioventricular block

first degree

0.113% incidence; 0.65-1.6% young adults ; 8.7%

trained athletes; 8% medical students [14]

Back Pain The lifetime prevalence has been estimated at anything

between 59% to 90% [15]

Bilirubinemia "hyperbilirubinemia " 3.1 -12.6 healthy infants [16]

Chest pain 2.21% USA; 2.1% placebo [17], [2]

Chills 35% in fever patients [18]

Congestive cardiac

failure

"Congestive Heart Failure/Cardiac

Dysfunction"

1.76% USA;2% age 40-59; 5% age 60-69; 10% over

70's [19]

Conjunctivitis "Keratoconjunctivitis sicca" [20]

Constipation 1.62% USA; 14.7% overall; 16% of children; and 15-

50% of the elderly; 3.8% placebo

[21], [2],

[22]

Coronary artery disease 4.85% USA (Coronary heart disease) [23]

Cough 0.00002% USA; 2.0% placebo [24], [2]

Dehydration 17.1% prevalence at admissions > 60 age; 6.7% elder;

0.55% in hospital [25], [26], [27]

Depression 5.3% USA; 0 placebo [28], [3]

Diarrhea "diarrhoea" 6.7% placebo; 100% USA [2]; [29]

Dizziness 4.8% orthostatic dizziness; 3.8% placebo; 1.7% or

5.9% placebo

[30]

[2]; [3]

Dry Skin 3.10% USA [31]

Dyspepsia 2.13% USA; 4.4% placebo; 5.9% placebo [32]; [2]; [3]

Normalized Adverse

event Other Terms Included Prevalence Reference

20

Dyspnea 6-27% adult Sharon E Mace, An Introduction to

Clinical Emergency Medicine p485

Dysuria 0.6% of women 0.1% of men [33]

Ecchymoses Unknown

Edema

"Generalized Edema"; "Periorbital

edema"; "Peripheral edema";

"Superficial Edema"; "Palmar-

plantar erthrodysesthesia"

1.6% USA (variable with country) [34]

Epistaxis 2-4% using placebo; 60% [35, 36]

Exfoliative Dermatitis 10% of childhood due to allergy to certain foods [37]

Fatigue "weakness;"Asthenia"; "loss of

strength" 4% asthenia, 2.54% Fatigue [38, 39]

Fever "Pyrexia"; "rigors" 2.3% placebo; <2% [2]

Flatulence 5.4% placebo [2]

Fluid Retention "Other Fluid Retention"; Unknown

Folliculitis "follicular rash" 20-25% of adult USA [40]

Hand-foot skin reaction "hand-foot syndrome" [38], [41]

Headache "Pain, headache"; 7% headache; 16.54% in USA; 23.5%; 15.6 placebo [38], [41], [3]; [2]

Hemorrhage

"ocular hemorrhage"; "petechiae";

"tumor hemorrhage"; "Cerebral

hemorrhage"; "CNS Bleeding";

"gastrointestinal hemorrhage"; "GI

Bleeding"

1.4% for 49 years or olders [42]

Hyperglycaemia estimated to be 121 per 1000 patients [43]

Hyperkeratosis 7.1% in Taiwain, affects 1 person per 2,000 European [44], [45]

Hypertension "Pulmonary Hypertension" 18.38% USA; 1.1%; 2.9% placebo; 16.1% (Miller

2005)

[46]; [3],

[30], [47]

Hypokalemia published incidence ranging from 4.6% (Greenfeld,

1995) to 19.7% (Miller, 2005) [48], [47]

Hypotension 10.7% for all, 20% elderly individuals (Postural

hypotension ) [49]

Hypothyroidism and/or

increased TSH 0.55% USA [50]

Increased lacrimation

Infection

"Infection (bacterial, fungal, non-

spec)"; "influenza"; "Urinary Tract

Infection"

1.1% placebo; 34%/13.9%/53.5% of

adults >20/woman/man; 3% children Urinary Tract

Infection;

[2]; [51]

Inhibit the cardiac QT

interval "prolonged QT interval"

9.2% in Caucasian patient with heart disease , while

6.5% US population reported a history of heart disease [52], [53]

Insomnia 11.76% USA; 3.4% placebo [54], [2]

Normalized Adverse

event Other Terms Included Prevalence Reference

21

Liver Enzymes

"ALP elevation"; "ALT elevation";

"ALT increased"; "AST

increased"; "Elevated

transaminases"; "gamma GT

increase"; "GGT increased"

11.6% among 40 years old woman in Tawain; more

than 15% of Greek blood donors exhibit elevated liver

enzymes

[55], [56]

Lymphopenia Unknown

Mucositis

"Mucositis/stomatitis"; "Mucosal

Inflammation (incl.

mucositis/stomatitis)"; "Mucosal

inflammation"; "mouth ulcer";

"Stomatitis"

19.5% control group [57]

Musculoskeletal Pain

"muscle cramps"; "Myalgia";

"Myalgia/limb pain"; "Pain in

extremity"; "bone pain"

2.21% back pain USA; 5.9% placebo; 4.7-74.4%

Spinak or back pain [58], [3];

Nausea "CML nausea" 7.4% placebo [2]

Neuropathy "Neuropathy-sensory"; "Sensory

neuropathy" 20 million Peripheral neuropathy USA => 6.7% [59]

Neutropenia " Febrile

neutropenia" ;"neutropenia"

in the US, it is estimated that there is approximately

one case per 100,000 population (Dale, unpublished

observations)

[60]

Night sweats 41% [61]

Pain

"Pain, other than abdominal";

"pharyngolaryngeal pain"; "Jaw

Pain"

2.9% placrbo [3]

Pericardial Effusion

"Pleural Effusion"; "pleural

effusion or pulmonary edema

hemorrhage"; "Pulmonary

Edema"; "pulmonary embolism";

"angioedema"

2 to 3.5% of all large effusion [62]

Proteinuria 2.43-3.54% [63]

Pruritus 1.3% placebo [2]

Rash

"Rash and Related";

"Rash/desquamation"; "Skin

Rash"; "vesiculobullous rash";

"acneiform rash"; maculopapular

rash

2.1% placebo [2]

Reduced LVEF

16.7% in older (63 ± 11 ) diabetic patients having

more peripheral arterial disease; while maximum

prevalence of diabete is 5%

[64, 65]

Normalized Adverse

event Other Terms Included Prevalence Reference

22

Respiratory Tract

Infection

"Rhinitis"; "sinusitis"; "sore

throat"; "upper respiratory tract

infection"; "Upper Respiratory

Tract Infection/Inflam";

"Interstitial lung disease

(interstitial pneumonia,

pneumonitis and alveolitis)";

nasopharyngitis; pneumonia

23.6% cold; 36% Flu; 15% Upper Respiratory

Infection; [66]

Skin Discoloration 1% in adult placebo; 3% in treatment patients in adults [67]

Thrombocytopenia 0.01% (Autoimmune Thrombocytopenia); 0.7%

(neonatal thrombocytopenia ) [68], [69]

Troponin I increase 3% in patients without pulmonary embolism [70]

Vomiting 5.6% placebo; 1.7% placebo Drugs.com[2]; dailymed

Weight Change

"Weight Decreased"; "Weight

Increase"; "weight increased";

weight loss; "GIST weight gain"

9.3% gain and 15% lose over 1 y [71]

*In this table, the adverse event “Asthenia” was merged with “Fatigue” because they are similar, and we further divided the adverse event “Pain”

into “Back Pain”, “Chest Pain” and “Pain” because their different incidences.

23

References of Suppl. Table 1:

1. WrongDiagnosis.com. Prevalence and Incidence of Abdominal Hernia. [cited; Available from:

http://www.wrongdiagnosis.com/a/abdominal_hernia/prevalence.htm.

2. Drugs.com. Axid Side Effects. [cited 2009 Feb 19]; Available from:

http://www.drugs.com/sfx/axid-side-effects.html.

3. dailymed.nlm.nih.gov. xyrem (sodium oxybate) solution [Jazz Pharmaceuticals, Inc.]

[cited; Available from: http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?id=1401.

4. Quigley, E.M., et al., Prevalence and management of abdominal cramping and pain: a

multinational survey. Aliment Pharmacol Ther, 2006. 24(2): p. 411-9.

5. WrongDiagnosis.com. Prevalence and Incidence of Acne. [cited; Available from:

http://www.wrongdiagnosis.com/a/acne/prevalence.htm.

6. Dobbins, H.M., et al., Interventions for alopecia areata. . Protocol) Cochrane Database of

Systematic Reviews,, 2003(4).

7. Webber, A.L. and J. Wood, Amblyopia: prevalence, natural history, functional effects and

treatment. Clin Exp Optom, 2005. 88(6): p. 365-75.

8. Salomao, S.R., et al., Prevalence and causes of visual impairment in low-middle income school

children in Sao Paulo, Brazil. Invest Ophthalmol Vis Sci, 2008. 49(10): p. 4308-13.

9. WrongDiagnosis.com. Prevalence and Incidence of Anemia. [cited; Available from:

http://www.wrongdiagnosis.com/a/anemia/prevalence.htm.

10. WrongDiagnosis.com. Prevalence and Incidence of Anorexia Nervosa. [cited; Available from:

http://www.wrongdiagnosis.com/a/anorexia_nervosa/prevalence.htm.

11. WrongDiagnosis.com. Prevalence Statistics for Types of Heart disease [cited; Available from:

http://www.wrongdiagnosis.com/h/heart_disease/prevalence-types.htm.

12. Choi, H.J., et al., Prevalence and clinical features of arthralgia/arthritis in healthy pregnant

women. Rheumatol Int, 2008. 28(11): p. 1111-5.

13. WrongDiagnosis.com. Prevalence and Incidence of Asthma. [cited; Available from:

http://www.wrongdiagnosis.com/a/asthma/prevalence.htm.

14. Levine, M.D. and S. Physician. Heart Block, First Degree. 2006 [cited; Available from:

http://emedicine.medscape.com/article/758322-overview.

15. Hermans, V., Research on work-related low back disorders. 2000: Institute for Occupational

Safety and Health, Brussels

16. Kavehmanesh, Z., et al., Prevalence of Readmission for Hyperbilirubinemia in Healthy Newborns.

Iran J Pediatr, 2008. 18(2): p. 130-136.

17. WrongDiagnosis.com. Prevalence and Incidence of Chest pain. [cited; Available from:

http://www.wrongdiagnosis.com/c/chest_pain/prevalence.htm.

18. Hoogendoorn, M., et al., [Predictive value of chills in patients presenting with fever to urgent care

department]. Ned Tijdschr Geneeskd, 2002. 146(3): p. 116-20.

19. WrongDiagnosis.com. Prevalence and Incidence of Congestive Heart Failure. [cited; Available

from: http://www.wrongdiagnosis.com/c/chest_pain/prevalence.htm.

20. Chhatwal, J. and P.V. Varughese, Neonatal conjunctivitis : a profile. Indian pediatrics, 1994.

30(11): p. 1357-1361.

21. WrongDiagnosis.com. Prevalence and Incidence of Constipation. [cited; Available from:

http://www.wrongdiagnosis.com/c/constipation/prevalence.htm.

22. Camilleri, M., et al., A placebo-controlled trial of prucalopride for severe chronic constipation. N

Engl J Med, 2008. 358(22): p. 2344-54.

23. WrongDiagnosis.com. Prevalence and Incidence of Coronary heart disease. [cited; Available

from: http://www.wrongdiagnosis.com/c/coronary_heart_disease/prevalence.htm.

24. WrongDiagnosis.com. Prevalence and Incidence of Whooping Cough. [cited; Available from:

http://www.wrongdiagnosis.com/w/whooping_cough/prevalence.htm.

24

25. Vivanti, A. Dehydration and older people…a drought within ? [cited; Available from:

http://www.health.qld.gov.au/hacc/Nutritional%20_%20Leaderslab/dehydration%20HACC%20

MOW%20-simplified.pdf.

26. Warren, J.L., et al., The burden and outcomes associated with dehydration among US elderly,

1991. Am J Public Health, 1994. 84(8): p. 1265-9.

27. Wakefield, B.J., et al., Risk factors and outcomes associated with hospital admission for

dehydration. Rehabil Nurs, 2008. 33(6): p. 233-41.

28. WrongDiagnosis.com. Prevalence and Incidence of Depression. [cited; Available from:

http://www.wrongdiagnosis.com/d/depression/prevalence.htm.

29. WrongDiagnosis.com. Prevalence and Incidence of Diarrhea. [cited; Available from:

http://www.wrongdiagnosis.com/d/diarrhea/prevalence.htm.

30. Wu, J.S., et al., Population-based study on the prevalence and correlates of orthostatic

hypotension/hypertension and orthostatic dizziness. Hypertens Res, 2008. 31(5): p. 897-904.

31. WrongDiagnosis.com. Prevalence and Incidence of Dry skin. [cited; Available from: Prevalence

and Incidence of Diarrhea.

32. WrongDiagnosis.com. Prevalence and Incidence of Nonulcer dyspepsia. [cited; Available from:

http://www.wrongdiagnosis.com/n/nonulcer_dyspepsia/prevalence.htm.

33. Patient.co.uk. Dysuria. [cited; Available from: http://www.patient.co.uk/showdoc/40000082/.

34. WrongDiagnosis.com. Statistics by Country for Edema [cited; Available from: Ecchymoses.

35. Benninger, M.S., Epistaxis and its relationship to handedness with use of intranasal steroid spray.

Ear Nose Throat J, 2008. 87(8): p. 463-5.

36. Gifford, T.O. and R.R. Orlandi, Epistaxis. Otolaryngol Clin North Am, 2008. 41(3): p. 525-36,

viii.

37. WrongDiagnosis.com. Prevalence and Incidence of Exfoliative dermatitis. [cited; Available

from: http://www.wrongdiagnosis.com/e/exfoliative_dermatitis/prevalence.htm.

38. Rosenzweig, P., S. Brohier, and A. Zipfel, The placebo effect in healthy volunteers: influence of

experimental conditions on the adverse events profile during phase I studies. Clin Pharmacol Ther,

1993. 54(5): p. 578-83.

39. Reeves, W.C., et al., Prevalence of chronic fatigue syndrome in metropolitan, urban, and rural

Georgia. Popul Health Metr, 2007. 5: p. 5.

40. US_Army_Center, f.H.P.p.m. Folliculitis. [cited; Available from: http://chppm-

www.apgea.army.mil/dhpw/Documents/part2KImbus11-13-07slide51-100.ppt.

41. WrongDiagnosis.com. Prevalence Statistics for Types of Pain. [cited; Available from:

http://www.wrongdiagnosis.com/p/pain/prevalence-types.htm.

42. Healey, P.R., et al., Optic disc hemorrhages in a population with and without signs of glaucoma.

Ophthalmology 1998. 105(2): p. 216-223.

43. Valero, M.A., et al., Clinical management of hyperglycaemic patients receiving total parenteral

nutrition. Clin Nutr, 1996. 15(1): p. 11-5.

44. Cebrian, M.E., et al., Chronic arsenic poisoning in the north of Mexico. Hum Toxicol, 1983. 2(1):

p. 121-33.

45. WrongDiagnosis.com. Prevalence and Incidence of Hyperkeratosis lenticularis perstans. [cited;

Available from:

http://www.wrongdiagnosis.com/h/hyperkeratosis_lenticularis_perstans/prevalence.htm.

46. WrongDiagnosis.com. Prevalence and Incidence of Hypertension. [cited; Available from:

http://www.wrongdiagnosis.com/h/hypertension/prevalence.htm.

47. Miller, K.K., et al., Medical findings in outpatients with anorexia nervosa. Arch Intern Med, 2005.

165(5): p. 561-6.

48. Greenfeld, D., et al., Hypokalemia in outpatients with eating disorders. Am J Psychiatry, 1995.

152(1): p. 60-3.

49. Mader, S.L., K.R. Josephson, and L.Z. Rubenstein, Low prevalence of postural hypotension

among community-dwelling elderly. JAMA, 1987. 258(11): p. 1511-4.

25

50. WrongDiagnosis.com. Prevalence Statistics for Types of Hypothyroidism [cited; Available from:

http://www.wrongdiagnosis.com/h/hypothyroidism/prevalence-types.htm.

51. WrongDiagnosis.com. Prevalence and Incidence of Urinary tract infections. [cited; Available

from: http://www.wrongdiagnosis.com/u/urinary_tract_infections/prevalence.htm.

52. Fonseca, F., et al., Prevalence of long QTc interval in methadone maintenance patients. Drug

Alcohol Depend, 2009. 99(1-3): p. 327-32.

53. CDC.gov, Prevalence of heart disease--United States, 2005. MMWR Morb Mortal Wkly Rep,

2007. 56(6): p. 113-8.

54. WrongDiagnosis.com. Prevalence and Incidence of Insomnia. [cited; Available from:

http://www.wrongdiagnosis.com/i/insomnia/prevalence.htm.

55. Lu, S.N., et al., Predicting the prevalence of antibody to hepatitis C virus (HCV) in a community

by the prevalence of elevated levels of alanine aminotransferase: a method to identify areas

endemic for HCV. Am J Trop Med Hyg, 2002. 67(2): p. 145-50.

56. Papatheodoridis, G.V., et al., High prevalence of elevated liver enzymes in blood donors:

associations with male gender and central adiposity. Eur J Gastroenterol Hepatol, 2007. 19(4): p.

281-7.

57. Ince, A., et al., The frequency and clinical characteristics of methotrexate (MTX) oral toxicity in

rheumatoid arthritis (RA): a masked and controlled study. Clin Rheumatol, 1996. 15(5): p. 491-4.

58. WrongDiagnosis.com. Prevalence and Incidence of Back pain. [cited; Available from:

http://www.wrongdiagnosis.com/b/back_pain/prevalence.htm.

59. Neurologychannel.com. Neuropathy. [cited; Available from:

http://www.neurologychannel.com/neuropathy/index.shtml.

60. neutropenia.ca. What is Neutropenia? [cited; Available from:

http://www.neutropenia.ca/about/index.html.

61. Medicinenet.com. Night Sweats. [cited; Available from:

http://www.medicinenet.com/night_sweats/article.htm.

62. Patient.co.uk. Pericardial Effusion. [cited; Available from:

http://www.patient.co.uk/showdoc/40024838/.

63. Pavkov, M.E., et al., Increasing incidence of proteinuria and declining incidence of end-stage

renal disease in diabetic Pima Indians. Kidney Int, 2006. 70(10): p. 1840-6.

64. Chareonthaitawee, P., et al., Prevalence and prognosis of left ventricular systolic dysfunction in

asymptomatic diabetic patients without known coronary artery disease referred for stress single-

photon emission computed tomography and assessment of left ventricular function. Am Heart J,

2007. 154(3): p. 567-74.

65. Currie, C.J. and J.R. Peters, Estimation of unascertained diabetes prevalence: different effects on

calculation of complication rates and resource utilization. Diabet Med, 1997. 14(6): p. 477-81.

66. WrongDiagnosis.com. Prevalence and Incidence of Upper Respiratory Infection. [cited;

Available from: http://www.wrongdiagnosis.com/u/uri/prevalence.htm adults.

67. Monodou, E., et al. incidence of Skin Discoloration Across Phase 3 Clinical Trails of

Emtricitabine (FTC) in adults. in XV International ADIS Conference. 2004. Bangkok, Thailand.

68. WrongDiagnosis.com. Prevalence and Incidence of Autoimmune Thrombocytopenia. [cited;

Available from:

http://www.wrongdiagnosis.com/a/autoimmune_thrombocytopenia/prevalence.htm.

69. Uhrynowska, M., K. Maslanka, and B. Zupanska, Neonatal thrombocytopenia: incidence,

serological and clinical observations. Am J Perinatol, 1997. 14(7): p. 415-8.

70. Yalamanchili, K., et al., Prevalence of increased cardiac troponin I levels in patients with and

without acute pulmonary embolism and relation of increased cardiac troponin I levels with in-

hospital mortality in patients with acute pulmonary embolism. Am J Cardiol, 2004. 93(2): p. 263-

4.

71. Jeffery, R.W., M.T. McGuire, and S.A. French, Prevalence and correlates of large weight gains

and losses. Int J Obes Relat Metab Disord, 2002. 26(7): p. 969-72.

J Biomed Inform. 2010 Jun;43(3):376-84. Epub 2010 May 1.

26

Supplementary Table 2 - The 17 normalized adverse events for 20 PKIs and their frequencies (F) in

percentage

PKI Name/Generic/Trade Company F (%) Adverse event Normalized

Adverse event Ref.

AMG-706 ("Motesanib diphosphate")

Amgen 70.0 diarrhea Diarrhea 1

AMG-706 ("Motesanib diphosphate")

Amgen 58.0 fatigue Fatigue 1

AMG-706 ("Motesanib diphosphate")

Amgen 43.0 headache Headache 1

AMG-706 ("Motesanib diphosphate")

Amgen 49.0 hypertension Hypertension 1

AMG-706 ("Motesanib diphosphate")

Amgen 40.0 nausea Nausea 1

BMS-387032/SNS-032 Sunesis 45.0 Abdominal Pain or cramping Abdominal Pain 5

BMS-387032/SNS-032 Sunesis 30.0 Anorexia Anorexia 5

BMS-387032/SNS-032 Sunesis 35.0 constipation Constipation 5

BMS-387032/SNS-032 Sunesis 30.0 Cough Cough 5

BMS-387032/SNS-032 Sunesis 30.0 Diarrhea Diarrhea 5

BMS-387032/SNS-032 Sunesis 90.0 Fatigue Fatigue 5

BMS-387032/SNS-032 Sunesis 35.0 nausea Nausea 5

CHIR-258/TKI-258 Novartis 17.1 Anorexia Anorexia 6

CHIR-258/TKI-258 Novartis 5.7 diarrhea Diarrhea 6

CHIR-258/TKI-258 Novartis 31.4 fatigue Fatigue 6

CHIR-258/TKI-258 Novartis 14.3 headache Headache 6

CHIR-258/TKI-258 Novartis 5.7 hypertension Hypertension 6

CHIR-258/TKI-258 Novartis 28.6 nausea Nausea 6

CHIR-258/TKI-258 Novartis 22.9 Vomiting Vomiting 6

CI-1033 Pfizer 19.1 abdominal pain Abdominal Pain 7

CI-1033 Pfizer 28.6 Anorexia Anorexia 7

CI-1033 Pfizer 40.0 Asthenia Asthenia 7

CI-1033 Pfizer 11.4 constipation Constipation 7

CI-1033 Pfizer 70.5 Diarrhea Diarrhea 7

CI-1033 Pfizer 24.8 Dry Skin Dry Skin 7

CI-1033 Pfizer 48.6 Mucositis Mucositis 7

CI-1033 Pfizer 48.6 nausea Nausea 7

CI-1033 Pfizer 11.4 pruritus Pruritus 7

CI-1033 Pfizer 46.7 Rash Rash 7

CI-1033 Pfizer 11.4 rhinitis Respiratory Tract

Infection 7

CI-1033 Pfizer 40.0 Vomiting Vomiting 7

CP-724714 Pfizer 33.3 Abdominal Pain Abdominal Pain 10

CP-724714 Pfizer 16.7 Anorexia Anorexia 10

CP-724714 Pfizer 43.3 Asthenia Asthenia 10

CP-724714 Pfizer 10.0 cough Cough 10

27

CP-724714 Pfizer 16.7 diarrhea Diarrhea 10

CP-724714 Pfizer 20.0 Drspnea Dyspnea 10

CP-724714 Pfizer 46.0 fatigue Fatigue 9

CP-724714 Pfizer 10.0 headache Headache 10

CP-724714 Pfizer 10.0 Hypotension Hypotension 10

CP-724714 Pfizer 60.0 nausea Nausea 10

CP-724714 Pfizer 20.0 pruritus Pruritus 10

CP-724714 Pfizer 33.3 Rash Rash 10

CP-724714 Pfizer 23.3 Respiratory Tract Infection Respiratory Tract

Infection 10

CP-724714 Pfizer 23.3 Vomiting Vomiting 10

Dasatinib/"Sprycel" Bristol-Myers

13.0 Chest Pain Abdominal Pain T

Dasatinib/"Sprycel" Bristol-Myers

19.0 Anorexia Anorexia T

Dasatinib/"Sprycel" Bristol-Myers

19.0 Asthenia Asthenia T

Dasatinib/"Sprycel" Bristol-Myers

14.0 constipation Constipation T

Dasatinib/"Sprycel" Bristol-Myers

28.0 Cough Cough T

Dasatinib/"Sprycel" Bristol-Myers

49.0 Diarrhea Diarrhea T

Dasatinib/"Sprycel" Bristol-Myers

32.0 Dyspnea Dyspnea T

Dasatinib/"Sprycel" Bristol-Myers

39.0 Fatigue Fatigue T

Dasatinib/"Sprycel" Bristol-Myers

40.0 Headache Headache T

Dasatinib/"Sprycel" Bristol-Myers

1.0 Pulmonary Hypertension Hypertension T

Dasatinib/"Sprycel" Bristol-Myers

16.0 Mucosal Inflammation (incl. mucositis/stomatitis)

Mucositis T

Dasatinib/"Sprycel" Bristol-Myers

34.0 Nausea Nausea T

Dasatinib/"Sprycel" Bristol-Myers

11.0 pruritus Pruritus T

Dasatinib/"Sprycel" Bristol-Myers

35.0 Skin Rash Rash T

Dasatinib/"Sprycel" Bristol-Myers

11.0 pneumonia Respiratory Tract

Infection T

Dasatinib/"Sprycel" Bristol-Myers

22.0 Vomiting Vomiting T

EKB-569 Wyeth 41.0 Anorexia Anorexia 11

EKB-569 Wyeth 66.0 asthenia Asthenia 11

EKB-569 Wyeth 10.0 constipation Constipation 11

EKB-569 Wyeth 93.0 diarrhea Diarrhea 11

EKB-569 Wyeth 17.0 Dry Skin Dry Skin 11

EKB-569 Wyeth 45.0 Stomatitis Mucositis 11

EKB-569 Wyeth 45.0 Nausea Nausea 11

EKB-569 Wyeth 14.0 pruritus Pruritus 11

EKB-569 Wyeth 83.0 rash Rash 11

28

EKB-569 Wyeth 28.0 Vomiting Vomiting 11

Erlotinib/"Tarceva" Genentech; OSI

11.0 Abdominal pain Abdominal Pain T

Erlotinib/"Tarceva" Genentech; OSI

52.0 Anorexia Anorexia T

Erlotinib/"Tarceva" Genentech; OSI

33.0 Cough Cough T

Erlotinib/"Tarceva" Genentech; OSI

54.0 Diarrhea Diarrhea T

Erlotinib/"Tarceva" Genentech; OSI

12.0 Dry Skin Dry Skin T

Erlotinib/"Tarceva" Genentech; OSI

41.0 Dyspnea Dyspnea T

Erlotinib/"Tarceva" Genentech; OSI

52.0 Fatigue Fatigue T

Erlotinib/"Tarceva" Genentech; OSI

17.0 Stomatitis Mucositis T

Erlotinib/"Tarceva" Genentech; OSI

33.0 Nausea Nausea T

Erlotinib/"Tarceva" Genentech; OSI

13.0 Pruritus Pruritus T

Erlotinib/"Tarceva" Genentech; OSI

75.0 Rash Rash T

Erlotinib/"Tarceva" Genentech; OSI

23.0 Vomiting Vomiting T

Gefitinib/ "Iressa" AstraZeneca 62.5 anorexia Anorexia P

Gefitinib/ "Iressa" AstraZeneca 62.5 asthenia Asthenia P

Gefitinib/ "Iressa" AstraZeneca 62.5 diarrhea Diarrhea P

Gefitinib/ "Iressa" AstraZeneca 62.5 dry skin Dry Skin P

Gefitinib/ "Iressa" AstraZeneca 17.5 dyspnea Dyspnea P

Gefitinib/ "Iressa" AstraZeneca 17.5 mouth ulcer Mucositis P

Gefitinib/ "Iressa" AstraZeneca 62.5 nausea Nausea P

Gefitinib/ "Iressa" AstraZeneca 62.5 pruritus Pruritus P

Gefitinib/ "Iressa" AstraZeneca 17.5 vesiculobullous rash Rash P

Gefitinib/ "Iressa" AstraZeneca 1.0 Interstitial lung disease (interstitial pneumonia, pneumonitis and alveolitis)

Respiratory Tract Infection

P

Gefitinib/ "Iressa" AstraZeneca 62.5 vomiting Vomiting P

GW-786034/ "Pazopanib" GlaxoSmithKline

3.9 Dyspnea

Dyspnea 12

GW-786034/ "Pazopanib" GlaxoSmithKline

3.9 Hypertension

Hypertension 12

GW-786034/ "Pazopanib" GlaxoSmithKline

3.9 pneumonia

Respiratory Tract Infection

12

Imatinib/"GLEEVEC" Novartis 62.5 Abdominal pain Abdominal Pain P

Imatinib/"GLEEVEC" Novartis 36.5 abdominal pain Abdominal Pain T

Imatinib/"GLEEVEC" Novartis 17.5 Anorexia Anorexia P

Imatinib/"GLEEVEC" Novartis 17.5 asthenia, weakness Asthenia P

Imatinib/"GLEEVEC" Novartis 11.4 constipation Constipation T

Imatinib/"GLEEVEC" Novartis 20.0 cough Cough T

Imatinib/"GLEEVEC" Novartis 45.4 Diarrhea Diarrhea T

Imatinib/"GLEEVEC" Novartis 17.5 Dyspnea Dyspnea P

29

Imatinib/"GLEEVEC" Novartis 38.8 fatigue Fatigue T

Imatinib/"GLEEVEC" Novartis 37.0 headache Headache T

Imatinib/"GLEEVEC" Novartis 17.5 hypokalemia Hypokalemia P

Imatinib/"GLEEVEC" Novartis 49.5 nausea Nausea T

Imatinib/"GLEEVEC" Novartis 17.5 pruritus Pruritus P

Imatinib/"GLEEVEC" Novartis 40.1 Rash and Related Rash T

Imatinib/"GLEEVEC" Novartis 11.4 sinusitis

Respiratory Tract Infection

T

Imatinib/"GLEEVEC" Novartis 22.5 vomiting Vomiting T

Lapatinib/"TYKERB" GlaxoSmithKline

65.0 Diarrhea Diarrhea T

Lapatinib/"TYKERB" GlaxoSmithKline

10.0 Dry skin Dry Skin T

Lapatinib/"TYKERB" GlaxoSmithKline

12.0 Dyspnea Dyspnea T

Lapatinib/"TYKERB" GlaxoSmithKline

15.0 Mucosal inflammation Mucositis T

Lapatinib/"TYKERB" GlaxoSmithKline

44.0 Nausea Nausea T

Lapatinib/"TYKERB" GlaxoSmithKline

28.0 Rash Rash T

Lapatinib/"TYKERB" GlaxoSmithKline

26.0 Vomiting Vomiting T

LY-333531 "Ruboxistaurin"

Eli Lilly and Company

24.4 Diarrhea

Diarrhea 13

MLN-518/ "Tandutinib" Millennium 42.5 Diarrhea Diarrhea 14

MLN-518/ "Tandutinib" Millennium 47.5 Nausea Nausea 14

MLN-518/ "Tandutinib" Millennium 37.5 Vomiting Vomiting 14

PKC-412/ "Midostaurin" Novartis 41.0 diarrhoea Diarrhea 3

PKC-412/ "Midostaurin" Novartis 41.0 fatigue Fatigue 3

PKC-412/ "Midostaurin" Novartis 41.0 nausea Nausea 3

PKC-412/ "Midostaurin" Novartis 41.0 vomiting Vomiting 3

PTK-787/ "Vatalanib" Novartis 100.

0 Fatigue Fatigue 15

PTK-787/ "Vatalanib" Novartis Hypertension Hypertension 15

Roscovitine/CYC-202/ "Seliciclib" Cyclacel

8.3 Anorexia Anorexia 16

Roscovitine/CYC-202/ "Seliciclib" Cyclacel

8.3 Fatigue Fatigue 16

Roscovitine/CYC-202/ "Seliciclib" Cyclacel

8.3 Hyponatraemia Hypokalemia 16

Roscovitine/CYC-202/ "Seliciclib" Cyclacel

8.3 Hypotension Hypotension 16

Roscovitine/CYC-202/ "Seliciclib" Cyclacel

8.3 Rash Rash 16

Sorafenib "Nexavar" Bayer 11.0 Pain, abdomen Abdominal Pain T

Sorafenib "Nexavar" Bayer 16.0 Anorexia Anorexia T

Sorafenib "Nexavar" Bayer 15.0 Constipation Constipation T

Sorafenib "Nexavar" Bayer 13.0 Cough Cough T

Sorafenib "Nexavar" Bayer 43.0 Diarrhea Diarrhea T

Sorafenib "Nexavar" Bayer 11.0 Dry skin Dry Skin T

30

Sorafenib "Nexavar" Bayer 14.0 Dyspnea Dyspnea T

Sorafenib "Nexavar" Bayer 37.0 Fatigue Fatigue T

Sorafenib "Nexavar" Bayer 10.0 Pain, headache Headache T

Sorafenib "Nexavar" Bayer 17.0 Hypertension Hypertension T

Sorafenib "Nexavar" Bayer 23.0 Nausea Nausea T

Sorafenib "Nexavar" Bayer 19.0 Pruritus Pruritus T

Sorafenib "Nexavar" Bayer 40.0 Rash/desquamation Rash T

Sorafenib "Nexavar" Bayer 16.0 Vomiting Vomiting T

Sunitinib malate/ "SUTENT"

Pfizer 33.0 Anorexia

Anorexia T

Sunitinib malate/ "SUTENT"

Pfizer 22.0 Asthenia

Asthenia T

Sunitinib malate/ "SUTENT"

Pfizer 20.0 constipation

Constipation T

Sunitinib malate/ "SUTENT"

Pfizer 40.0 Diarrhea

Diarrhea T

Sunitinib malate/ "SUTENT"

Pfizer 15.0 Hypertension

Hypertension T

Sunitinib malate/ "SUTENT"

Pfizer 29.0 Mucositis/stomatitis

Mucositis T

Sunitinib malate/ "SUTENT"

Pfizer 14.0 rash

Rash T

ZD-6474 ("Vandetanib" "Zactima")

AstraZeneca 13.0 anorexia

Anorexia 2

ZD-6474 ("Vandetanib" "Zactima")

AstraZeneca 37.7 diarrhea

Diarrhea 2

ZD-6474 ("Vandetanib" "Zactima")

AstraZeneca 18.2 fatigue

Fatigue 2

ZD-6474 ("Vandetanib" "Zactima")

AstraZeneca 18.2 hypertension

Hypertension 2

ZD-6474 ("Vandetanib" "Zactima")

AstraZeneca 19.5 nausea

Nausea 2

ZD-6474 ("Vandetanib" "Zactima")

AstraZeneca 10.4 maculopapular rash

Rash 2

T: Thompson PDR (online); P: PharmGKB

J Biomed Inform. 2010 Jun;43(3):376-84. Epub 2010 May 1.

31

Reference of Suppl. Table 2:

1. S. I. Sherman, M. J. Schlumberger, J. Droz, M. Hoffmann, L. Wirth, L. Bastholt, R. G.

Martins, L. Licitra, Y. Shi, D. E. Stepan Initial results from a phase II trial of

motesanib diphosphate (AMG 706) in patients with differentiated thyroid cancer

(DTC). Journal of Clinical Oncology, 2007 ASCO Annual Meeting Proceedings Part

I. Vol 25, No. 18S (June 20 Supplement), 2007: 6017

2. Holden SN, Eckhardt SG, Basser R, et al: Clinical evaluation of ZD6474, an orally

active inhibitor of VEGF and EGF receptor signaling, in patients with solid,

malignant tumors. Ann Oncol 16:1391-1397, 2005

3. MJ Millward*, C House, D Bowtell, L Webster, IN Olver, M Gore, M Copeman, K

Lynch, A Yap, Y Wang, PS Cohen and J Zalcberg. The multikinase inhibitor

midostaurin (PKC412A) lacks activity in metastatic melanoma: a phase IIA clinical

and biologic study. British Journal of Cancer (2006) 95, 829 – 834.

4. R. A. Soo, E. McKeegan, C. S. Chen, C. H. Thng, T. S. Koh, D. Laird, K. Zhang, C. I.

Wong, N. Gupta, B. C. Goh. The effect of varying doses of ABT-869 on biomarkers

of angiogenesis and their correlation with pharmacodynamic outcome. J Clin Oncol

26: 2008 (May 20 suppl; abstr 14535)

5. Elisabeth I. Heath & Keith Bible & Robert E. Martell & Daniel C. Adelman & Patricia

M. LoRusso. A phase 1 study of SNS-032 (formerly BMS-387032), a potent inhibitor

of cyclin-dependent kinases 2, 7 and 9 administered as a single oral dose and weekly

infusion in patients with metastatic refractory solid tumors. Invest New Drugs (2008)

26:59–65.

6. Debashis Sarker1, Rhoda Molife1, T.R. Jeffrey Evans2, Maryon Hardie2, Cheryl

Marriott1, Priska Butzberger-Zimmerli2, Rosemary Morrison2, Judith A. Fox3, Carla

Heise3, Sharianne Louie3, Natasha Aziz3, Felix Garzon3, Glenn Michelson3, Ian R.

Judson1, Dalal Jadayel1, Edgar Braendle3 and Johann S. de Bono1 . A Phase I

Pharmacokinetic and Pharmacodynamic Study of TKI258, an Oral, Multitargeted

Receptor Tyrosine Kinase Inhibitor in Patients with Advanced Solid Tumors. Clinical

Cancer Research 14, 2075-2081, April 1, 2008.

7. Susana Campos, Oday Hamid, Michael V. Seiden, Amit Oza, Marie Plante, Ronald K.

Potkul, Peter F. Lenehan, Eric P. Kaldjian, Mary L. Varterasian, Cheryl Jordan,

Claudie Charbonneau, and Hal Hirte. Multicenter, Randomized Phase II Trial of Oral

CI-1033 for Previously Treated Advanced Ovarian Cancer. JOURNAL OF

CLINICAL ONCOLOGY VOLUME 23 NUMBER 24 AUGUST 20 2005.

8. Tracy Hampton. Arthritis Clinical Trial Results Revealed. JAMA Vol. 297 No. 1,

January 3, 2007

9. P. N. Munster, M. Mita, C. Britten, S. Minton, S. Moulder, D. Noe, B. Roedig, L.

Denis, D. Slamon and A. Tolcher. Phase I and pharmacokinetic (PK) Study of CP-

724,714, an oral human epidermal growth factor receptor-2 (HER-2) selective

tyrosine kinase inhibitor. Journal of Clinical Oncology, 2004 ASCO Annual Meeting

Proceedings (Post-Meeting Edition). Vol 22, No 14S (July 15 Supplement), 2004:

3082

32

10. Pamela N. Munster1, Carolyn D. Britten2, Monica Mita3, Karen Gelmon4, Susan E.

Minton1, Stacy Moulder1, Dennis J. Slamon2, Feng Guo5, Stephen P. Letrent5,

Louis Denis5 and Anthony W. Tolcher3 . First Study of the Safety, Tolerability, and

Pharmacokinetics of CP-724,714 in Patients with Advanced Malignant Solid HER2-

Expressing Tumors. Clinical Cancer Research 13, 1238-1245, February 15, 2007. doi:

10.1158/1078-0432.CCR-06-1539

11. Charles Erlichman, Manuel Hidalgo, Joseph P. Boni, Patricia Martins, Susan E. Quinn,

Charles Zacharchuk, Peter Amorusi, Alex A. Adjei, Eric K. Rowinsky . Phase I Study

of EKB-569, an Irreversible Inhibitor of the Epidermal Growth Factor Receptor, in

Patients With Advanced Solid Tumors. Journal of Clinical Oncology, Vol 24, No 15

(May 20), 2006: pp. 2252-2260

12. N. Altorki, M. Guarino, P. Lee, H. I. Pass, E. Filip, T. Bauer, D. Roychowdhury, T.

Zaks, L. Ottesen, D. Yankelevitz. Preoperative treatment with pazopanib

(GW786034), a multikinase angiogenesis inhibitor in early-stage non-small cell lung

cancer (NSCLC): A proof-of-concept phase II study. J Clin Oncol 26: 2008 (May 20

suppl; abstr 7557)

13. The PKC-DRS Study Group. The Effect of Ruboxistaurin on Visual Loss in Patients

With Moderately Severe to Very Severe Nonproliferative Diabetic Retinopathy.

Diabetes 54:2188-2197, 2005.

14. Daniel J De Angelo, Richard M Stone, Mark L Heaney, Stephen D Nimer, Ronald L

Paquette, Rebecca B Klisovic, Michael A Caligiuri, Michael R Cooper, Jean-Michel

Lecerf, Michael D Karol, Shihong Sheng, Nick Holford, Peter T Curtin, Brian J

Druker and Michael C Heinrich. Phase I clinical results with tandutinib (MLN518), a

novel FLT3 antagonist, in patients with acute myelogenous leukemia or high-risk

myelodysplastic syndrome: safety, pharmacokinetics, and pharmacodynamics.

doi:10.1182/blood-2006-02-005702. Prepublished online Aug 15, 2006

15. R. DeBoer, S. A. Grimm, J. Chandler, L. Gallot, S. Kennedy, K. Burns, L. Rice, C.

Cabreza, K. Muro, J. J. Raizer. A phase II trial of PTK-787 (PTK/ZK) in recurrent or

progressive meningiomas. J Clin Oncol 26: 2008 (May 20 suppl; abstr 2060)

16. C Benson, J White, J De Bono, A O’Donnell, F Raynaud, C Cruickshank, H McGrath,

M Walton, P Workman, S Kaye, J Cassidy, A Gianella-Borradori, I Judson, and C

Twelves. A phase I trial of the selective oral cyclin-dependent kinase inhibitor

seliciclib (CYC202; R-Roscovitine), administered twice daily for 7 days every 21

days.British Journal of Cancer (2007) 96, 29-37.

J Biomed Inform. 2010 Jun;43(3):376-84. Epub 2010 May 1.

33

Supplementary Table 3 - The 41 prioritized AE-KT pairs identified using prioritized association score. The evaluated significant (p ≤ 0.05) pairs are

highlighted in red for their p-values of co-occurrence. The PKIs of each kinase are obtained from the paper of Karaman et al (2008).

Computation Prediction Literature Evaluation Kinome Map (PKI in Brown bold)

Target AE Score p q-value Fisher.p #KT #AE #KT &

AE Drugs(Kd<100nM)

EGFR Asthenia 0.13 <0.01 <0.01 < 0.001 11,087 4,401 9 CI.1033, CP.724714, EKB.569, Gefitinib

EGFR Diarrhea 0.33 <0.01 <0.01 < 0.001 10,908 68,235 133 AMG.706, CI.1033, CP.724714, EKB.569, Erlotinib, Gefitinib, Lapatinib, ZD.6474

EGFR Rash 0.2 <0.01 <0.01 < 0.001 10,908 15,264 181 CI.1033, CP.724714, EKB.569, Erlotinib, Gefitinib, Lapatinib, ZD.6474

ERBB2 Asthenia 0.12 <0.01 <0.01 < 0.001 11,087 4,401 11 CI.1033, CP.724714

ERBB2 Rash 0.16 <0.01 <0.01 < 0.001 11,087 15,264 33 CI.1033, CP.724714, Lapatinib

KIT Diarrhea 0.12 0.26 0.21 < 0.001 22,406 68,235 195 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, MLN.518, Sorafenib, Sunitinib

ERBB2 Nausea 0.25 <0.01 <0.01 0.01 11,087 35,735 32 CI.1033, CP.724714, Lapatinib

VEGFR2 Fatigue 0.13 0.02 0.21 0.05 582 44,331 4 AMG.706, CHIR.258.TKI.258, PTK.787, Sorafenib

ERBB4 Asthenia 0.11 0.01 0.15 0.17 811 4,401 1 CI.1033, Dasatinib, EKB.569

LOK Diarrhea 0.2 0.03 0.21 0.22 1,595 68,235 8 AMG.706, Sunitinib, ZD.6474

ERBB2 Diarrhea 0.23 <0.01 <0.01 0.23 11,087 68,235 45 CI.1033, CP.724714, Lapatinib

EGFR Fatigue 0.13 0.06 0.21 0.27 11,087 44,331 29 AMG.706, CP.724714, Erlotinib, ZD.6474

LOK Nausea 0.14 0.12 0.21 0.35 1,595 35,735 4 AMG.706, ZD.6474

ERBB4 Rash 0.16 <0.01 <0.01 0.48 811 15,264 1 CI.1033, Dasatinib, EKB.569, Lapatinib

EGFR Nausea 0.33 <0.01 <0.01 0.49 11,087 35,735 21 AMG.706, CI.1033, CP.724714, EKB.569, Erlotinib, Gefitinib, Lapatinib, ZD.6474

ERBB4 Diarrhea 0.24 0.01 0.15 0.56 811 68,235 3 CI.1033, Dasatinib, EKB.569, Lapatinib

PDGFRB Fatigue 0.17 0.02 0.21 0.56 2,582 44,331 6 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, PTK.787, Sorafenib, ZD.6474

PDGFRB Diarrhea 0.12 0.24 0.21 0.58 2,582 68,235 9 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, MLN.518, Sorafenib, Sunitinib, ZD.6474

FLT1 Fatigue 0.15 0.08 0.21 0.64 435 44,331 1 AMG.706, CHIR.258.TKI.258, PTK.787, Sorafenib

PDGFRA Diarrhea 0.12 0.19 0.21 0.64 2,427 68,235 8 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, MLN.518, Sorafenib, Sunitinib

34

ERBB4 Nausea 0.25 0.02 0.21 0.78 811 35,735 1 CI.1033, Dasatinib, EKB.569, Lapatinib

PDGFRA Fatigue 0.16 0.01 0.15 0.82 2,427 44,331 4 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, PTK.787, Sorafenib

ZAK Fatigue 0.12 0.13 0.21 0.95 1,998 44,331 4 AMG.706, Dasatinib, Sorafenib

KIT Fatigue 0.17 0.03 0.21 0.97 22,406 44331 40 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, PTK.787, Sorafenib

ERBB2 Vomiting 0.14 <0.01 <0.01 0.99 11,087 45,907 15 CI.1033, CP.724714, Lapatinib

CDK11 Nausea 0.12 0.2 0.21 1 28 35,735 0

CSF1R Fatigue 0.16 0.03 0.21 1 4,915 44,331 1 AMG.706, CHIR.258.TKI.258, Dasatinib, Imatinib, PTK.787, Sorafenib

DDR1 Diarrhea 0.12 0.15 0.21 1 121 68,235 0 AMG.706, Imatinib, Sorafenib, ZD.6474

DDR1 Fatigue 0.15 0.03 0.21 1 121 44,331 0 AMG.706, Imatinib, Sorafenib, ZD.6474

EGFR Vomiting 0.18 <0.01 <0.01 1 11,087 45,907 14 CI.1033, CP.724714, EKB.569, Erlotinib, Gefitinib, Lapatinib

ERBB4 Vomiting 0.14 0.01 0.15 1 811 45,907 0 CI.1033, Dasatinib, EKB.569, Lapatinib

FLT4 Fatigue 0.13 0.1 0.21 1 82 44,331 0 AMG.706, Sorafenib

FRK Fatigue 0.13 0.05 0.21 1 131 44,331 0 AMG.706, Dasatinib

LCK Diarrhea 0.12 0.2 0.21 1 2,370 68,235 1 AMG.706, EKB.569, Imatinib, ZD.6474

LYN Diarrhea 0.11 0.16 0.21 1 1,545 68,235 0 Dasatinib

MKNK2 Nausea 0.13 0.13 0.21 1 21 35,735

RIPK2 Diarrhea 0.16 0.11 0.21 1 93 68,235 0 CI.1033, ZD.6474

RIPK2 Nausea 0.11 0.25 0.21 1 93 35,735 0 CI.1033, ZD.6474

EGFR Edema 0.15 0.02 0.21 1 11,087 100,00

0 22 Dasatinib, Lapatinib

ERBB2 Edema 0.13 0.03 0.21 1 11,087 100,00

0 3 Lapatinib

ERBB4 Edema 0.12 0.08 0.21 1 811 100,00

0 0 Dasatinib, Lapatinib

-: no primary targets are given in the paper of Karaman et al (2008).

35

Supplementary Table 4 - The manually verification resulted in 78 out of 105 significant (unadjusted p-value≤0.05) AE-KT pairs

that were identified using text mining in PubMed. Adjusted p-value ≤ 5% results followed by those with unadjusted p-value ≤ 5% but

adj. p-value > 5%.

KT AE #co-

occurrence fisher

.p adjusted.p #KT #SE

Method

Verification Annotation

ACVRL1 Hypertension 17 0 0 158 100,000

AKT1 Hypertension 54 0 0 3,866 100,000

ALK Hypertension 21 0 0 2,025 100,000

CSK Hypertension 28 0 0 1,523 100,000

CSK Respiratory Tract Infection 37 0 0 1,523 100,000

DLK Hypertension 7 0 0 229 100,000

EGFR Asthenia 9 0 0 10,928 4,401 1

EGFR Diarrhea 133 0 0 10,928 68,235 1+2 [1], [2], [3]

EGFR Dry Skin 11 0 0 10,928 3,220

EGFR Hypertension 198 0 0 10,928 100,000

EGFR Rash 181 0 0 10,928 15,264 1+2 [3], [2]

ERBB2 Asthenia 11 0 0 11,109 4,401 1 [4]

ERBB2 Rash 33 0 0 11,109 15,264 1 [5], [6]

FES Fatigue 83 0 0 2,385 44,331

FES Rash 7 0 0 2,385 15,264

FGFR2 Mucositis 6 0 0 1,251 4,500

FGR Hypertension 33 0 0 616 100,000

FLT1 Hypertension 27 0 0 437 100,000

IGF1R Hypertension 32 0 0 3,207 100,000

INSR Hypertension 15 0 0 162 100,000

KIT Abdominal Pain 89 0 0 22,433 46,779

KIT Constipation 34 0 0 22,433 13,434

KIT Diarrhea 196 0 0 22,433 68,235 1 [6]

KIT Hypertension 159 0 0 22,433 100,000

KT AE # co-

occurrence fisher.p adjusted.p

#KT #SE Method Verification Annotation

36

KIT Rash 31 0 0 22,433 15,264

KIT Respiratory Tract Infection 342 0 0 22,433 100,000

LOK Abdominal Pain 15 0 0 1,598 46,779

LOK Hypertension 18 0 0 1,598 100,000

LOK Respiratory Tract Infection 21 0 0 1,598 100,000

MEK1 Hypertension 29 0 0 2,448 100,000

MLCK Hypertension 12 0 0 821 100,000

MUSK Cough 10 0 0 1,521 28,493

MUSK Dyspnea 11 0 0 1,521 27,119

MUSK Nausea 13 0 0 1,521 35,735

MUSK Vomiting 49 0 0 1,521 45,907

MYLK2 Hypertension 89 0 0 3,604 100,000

PDGFRA Hypertension 43 0 0 2,428 100,000

PDGFRB Hypertension 64 0 0 2,583 100,000

PRKCE Hypertension 9 0 0 588 100,000

PTK2B Hypertension 8 0 0 231 100,000

RET Constipation 17 0 0 3,762 13,434 2 [7], [8], [9]

RET Hypertension 36 0 0 3,762 100,000 2 [10], [11]

SLK Abdominal Pain 4 0 0 160 46,779

TIE2 Hypertension 14 0 0 577 100,000

VEGFR2 Hypertension 10 0 0 584 100,000

ALK Fatigue 11 0.01 0.595 2,025 44,331

DDR2 Hypertension 3 0.01 0.595 71 100,000

ERBB2 Nausea 32 0.01 0.595 11,109 35,735 1 [12]

FER Constipation 3 0.01 0.595 685 13,434

FER Mucositis 2 0.01 0.595 685 4,500

LOK Rash 5 0.01 0.595 1,598 15,264

KT AE # co-

occurrence fisher.p adjusted.p

#KT #SE Method Verification Annotation

37

PDGFRA Rash 6 0.01 0.595 2,428 15,264

SLK Hypertension 4 0.01 0.595 160 100,000

TEC Anorexia 6 0.01 0.595 1,705 20,259

ACVRL1 Dyspnea 2 0.02 1 158 27,119

EGFR Mucositis 7 0.02 1 10,928 4,500 2 [13]

ERBB2 Mucositis 7 0.02 1 11,109 4,500

FER Dyspnea 4 0.02 1 685 27,119

LOK Pruritus 4 0.02 1 1,598 12,739

MST1 Dry Skin 1 0.02 1 129 3,220

PHKG2 Fatigue 1 0.02 1 9 44,331

PRKCD Hypertension 9 0.02 1 783 100,000

PRKX Constipation 1 0.02 1 25 13,434

MEK2 Dry Skin 1 0.03 1 205 3,220

MUSK Fatigue 8 0.03 1 1,521 44,331

PRKCH Hypertension 1 0.03 1 6 100,000

TEC Dry Skin 2 0.03 1 1,705 3,220

ANKK1 Fatigue 1 0.04 1 18 44,331

EGFR Pruritus 13 0.04 1 10,928 12,739

FES Anorexia 6 0.04 1 2,385 20,259

LOK Cough 6 0.04 1 1,598 28,493

TNK1 Respiratory Tract Infection 1 0.04 1 8 100,000

FES Respiratory Tract Infection 19 0.05 1 2,385 100,000

MEK2 Asthenia 1 0.05 1 205 4,401

STK36 Respiratory Tract Infection 1 0.05 1 9 100,000

VEGFR2 Fatigue 4 0.05 1 584 44,331 1

[14], [15], [16], [17]

CIT Diarrhea 26 0 0 2185 68235 2 Rejected CIT=city in address

KT AE # co-

occurrence fisher.p adjusted.p

#KT #SE Method Verification Annotation

38

GAK Anorexia 3 0 0 126 20259 2 Rejected Gak as institute name

MET Abdominal Pain 256 0 0 66248 46779 Rejected "Methods" "metabolic equivalents"

MET Anorexia 249 0 0 66248 20259 Rejected "Methods" "metabolic equivalents"

MET Asthenia 29 0 0 66248 4401 Rejected "Methods" "metabolic equivalents"

MET Constipation 123 0 0 66248 13434 Rejected "Methods" "metabolic equivalents"

MET Cough 170 0 0 66248 28493 Rejected "Methods" "metabolic equivalents"

MET Diarrhea 387 0 0 66248 68235 Rejected "Methods" "metabolic equivalents"

MET Dyspnea 143 0 0 66248 27119 Rejected "Methods" "metabolic equivalents"

MET Fatigue 471 0 0 66248 44331 Rejected "Methods" "metabolic equivalents"

MET Headache 457 0 0 66248 45968 Rejected "Methods" "metabolic equivalents"

MET Hypertension 1153 0 0 66248 100,000 Rejected "Methods" "metabolic equivalents"

MET Nausea 388 0 0 66248 35735 Rejected "Methods" "metabolic equivalents"

MET Rash 113 0 0 66248 15264 Rejected "Methods" "metabolic equivalents"

MET Respiratory Tract Infection 825 0 0 66248 100,000 Rejected "Methods" "metabolic equivalents"

MET Vomiting 357 0 0 66248 45907 Rejected "Methods" "metabolic equivalents"

YES Headache 37 0 0 5212 45968 Rejected as" Yes/no"

YES Hypertension 81 0 0 5212 100,000 Rejected as" Yes/no"

ZAK Respiratory Tract Infection 21 0 0 1999 100,000 Rejected ZAk as author name

FER Hypertension 18 0 0 685 100,000 Rejected "Fer" as author name

SRC Hypertension 170 0 0 20648 100,000 Rejected "scleroderma renal crisis "

MET Mucositis 27 0.01 0.595 66248 4500 Rejected "Methods" "metabolic equivalents"

ZAK Hypertension 18 0.02 1 1999 100,000 Rejected ZAk as author name

GAK Respiratory Tract Infection 3 0.03 1 126 100,000 Rejected Gak as institute name

GAK Abdominal Pain 2 0.04 1 126 46779 Rejected Gak as institute name

GAK Headache 2 0.04 1 126 45968 Rejected Gak as institute name

TEC Respiratory Tract Infection 15 0.04 1 1705 100,000 Rejected

"toxigenic Escherichia coli " "total eosinophil counts " etc.

YES Dyspnea 13 0.04 1 5212 27119 Rejected as" Yes/no"

YES Respiratory Tract Infection 37 0.05 1 5212 100,000 Rejected as" Yes/no"

39

References of Suppl. Table 4:

1. Herbst, R.S., et al., Selective oral epidermal growth factor receptor tyrosine kinase inhibitor ZD1839

is generally well-tolerated and has activity in non-small-cell lung cancer and other solid tumors:

results of a phase I trial. J Clin Oncol, 2002. 20(18): p. 3815-25.

2. Erlichman, C., et al., Phase I study of EKB-569, an irreversible inhibitor of the epidermal growth

factor receptor, in patients with advanced solid tumors. J Clin Oncol, 2006. 24(15): p. 2252-60.

3. Ito, Y., et al., Does lapatinib, a small-molecule tyrosine kinase inhibitor, constitute a breakthrough in

the treatment of breast cancer? Breast Cancer, 2007. 14(2): p. 156-62.

4. Gasparini, G., et al., Gefitinib (ZD1839) combined with weekly epirubicin in patients with metastatic

breast cancer: a phase I study with biological correlate. Ann Oncol, 2005. 16(12): p. 1867-73.

5. Di Leo, A., et al., Phase III, double-blind, randomized study comparing lapatinib plus paclitaxel with

placebo plus paclitaxel as first-line treatment for metastatic breast cancer. J Clin Oncol, 2008. 26(34):

p. 5544-52.

6. Schlumberger, M.J., et al., Phase II study of safety and efficacy of motesanib in patients with

progressive or symptomatic, advanced or metastatic medullary thyroid cancer. J Clin Oncol, 2009.

27(23): p. 3794-801.

7. Edery, P., et al., Long segment and short segment familial Hirschsprung's disease: variable clinical

expression at the RET locus. J Med Genet, 1994. 31(8): p. 602-6.

8. Svensson, P.J., et al., Phenotypic variation in a family with mutations in two Hirschsprung-related

genes (RET and endothelin receptor B). Hum Genet, 1998. 103(2): p. 145-8.

9. King, S.K., B.R. Southwell, and J.M. Hutson, An association of multiple endocrine neoplasia 2B, a

RET mutation; constipation; and low substance P-nerve fiber density in colonic circular muscle. J

Pediatr Surg, 2006. 41(2): p. 437-42.

10. Dora, J.M., et al., Pancreatitis as the first manifestation of multiple endocrine neoplasia type 2A. Arq

Bras Endocrinol Metabol, 2008. 52(8): p. 1332-6.

11. Zhang, Z., et al., A common RET variant is associated with reduced newborn kidney size and

function. J Am Soc Nephrol, 2008. 19(10): p. 2027-34.

12. Medina, P.J. and S. Goodin, Lapatinib: a dual inhibitor of human epidermal growth factor receptor

tyrosine kinases. Clin Ther, 2008. 30(8): p. 1426-47.

13. Tejwani, A., et al., Increased risk of high-grade dermatologic toxicities with radiation plus epidermal

growth factor receptor inhibitor therapy. Cancer, 2009. 115(6): p. 1286-99.

14. Miyanaga, N. and H. Akaza, [Sorafenib(Nexavar)]. Gan To Kagaku Ryoho, 2009. 36(6): p. 1029-33.

15. Fury, M.G., et al., A Phase II study of SU5416 in patients with advanced or recurrent head and neck

cancers. Invest New Drugs, 2007. 25(2): p. 165-72.

16. Millward, M.J., et al., The multikinase inhibitor midostaurin (PKC412A) lacks activity in metastatic

melanoma: a phase IIA clinical and biologic study. Br J Cancer, 2006. 95(7): p. 829-34.

17. Stadler, W.M., et al., A randomized Phase II trial of the antiangiogenic agent SU5416 in hormone-

refractory prostate cancer. Clin Cancer Res, 2004. 10(10): p. 3365-70.