Transcript Biological and clinical heterogeneity of breast cancer

Biological and clinical
heterogeneity of
breast cancer
Antonio Frassoldati
Oncologia Clinica - Ferrara
Heterogeneity, …
that is, what we
have tried to ignore
Facing tumor heterogeneity
Looking changes by re-biopsy?
Mean discondance 29,7%
Kasraw, Curr Oncol Rep 2011
Biological heterogeneity is the key
problem in precision medicine
Santarpia, The Oncologist, 2016
Implication of the concept that tumors are
composed of evolving clones
• Existence of clonal genotypes (i.e., not all mutations occur
in the same cells)
• Expansion and decline of clonal populations over time
• Existence of internal spatial variation in tumor composition
• Partial tumor responses to therapy and the emergence of
drug-resistant malignant cells
• Seeding of metastatic cells from subclones (which may be
rare or common in the originating population)
• Absence of an observable clonal structure based on
genome aberrations in some cancers
• Existence of neutral clonal relationships (e.g., arising from
random genetic drift) without discernible phenotypic
consequences.
Mutation, Selection, and Drift are the Three
Basic Processes Shaping Cancer Evolution
Clonal selection
Multiple intratumoral subclones harboring different driver mutations,
displaying distinct phenotypes, and evolving with branched
phylogenies were identified; spatial constraints most likely limit
clonal competition to the immediately neighboring subclones
Mutation are the
prerequisite for evolution,
but their rates, the
genomic regions that are
prone to mutagenesis, and
the timing when particular
mutagenic processes
operate can vary
significantly between but
also within individual
cancers.
Genetic drift
each cell in a newly
generated cancer
subclone has a certain
probability of dying as a
result of random factors
and occasionally all
cells of a small
subclone die, even if
this clone harbors a
highly beneficial
mutation.
Lipinski, Trends in Cancer 2016
Mutation rate increases with increase in tumor size
Lipinski, Trends in Cancer 2016
Major cancer susceptibility loci
identified in breast cancer
Santarpia, The Oncologist, 2016
Somatic mutations in breast cancer and molecular subtypes
according to COSMIC database
Santarpia, The Oncologist, 2016
Significantly Mutated Genes (SMG) and correlations
with genomic and clinical features
The Cancer Genome Atlas Network, Nature 2012
Frequencies of the most commonly mutated cancerrelated genes across breast cancer subtypes
according to the Cancer Genome Atlas
Santarpia, The Oncologist, 2016
Frequencies of the most commonly mutated cancerrelated genes across breast cancer subtypes
according to the Cancer Genome Atlas
Santarpia, The Oncologist, 2016
The dark side of genome
• Besides the importance of variants in protein-coding
regions, the majority of the alterations occur in
noncoding portions of the genome.
• Like somatic variants, germline noncoding affects
gene expression through several mechanisms (e.g.,
promoter mutations, single-nucleotide polymorphisms in
enhancers and noncoding RNAs and their binding sites, and
variants in introns)
• Breast cancer seems to harbor more alterations in
the noncoding regions compared with other tumors
• Noncoding driver mutations/alterations remains
crucial to enable therapeutic approaches that target
the specific linked proteins.
Cancer Evolution Features
• Tumor
microenvironmental
features, such as blood
vessel densities or
immune cell infiltrates,
can be responsible for
some relevant selection
pressures.
• Cancer cells colonizing
metastatic sites are also
likely to encounter altered
selective landscapes.
Lipinski, Trends in Cancer 2016
Clinical consequences
of biological heterogeneity
Santarpia, The Oncologist, 2016
Cancer as a society of neoplastic cells
(an anthropologic view…)
• Several variables have to be considered
– The composition of society
• Young or elderly people prevalence, foreign people, families
or singles,….
– The time point of observation
• Which idea dominates the society at that point
• Trying to understand the film looking at a picture
– The space dimension of society
• Site and space and time effects on society evolution
– The drivers of society
• Matriarcal or patriarcal society, culture, politics, religions….
– The relationships between societies
• Guests and hosts
Which effects of tumor heterogenity for the clinicians?
Limits or opportunities?
• Knowledge and prognostication of the disease
behavior (aggressiveness, specific site of
localization, …)
• Prediction of response to therapies
• Possibility to adapt disease monitoring and
treatment
• Possibility to modulate treatment based on
tumor characteristics and their changes
Inter-tumor heterogeneity and
response to therapy
Co-expression of HR, or p95 positivity in HER2 positive BC
reduce the probability of response to antiHER2 agents
Guarneri, JCO 2012
Response to primary therapy in HER2pos
BC by PIK3CA mut status
Loibl, Ann Oncol 2016
Mutation load affects PFS Benefit With Everolimus
in ER+ BC resistant to AIs
Subgroup
N
Events (%)
Median
PFS (d)
EVE: WT
PBO: WT
EVE: Single
PBO: Single
EVE: multiple
PBO: multiple
43
18
76
35
38
17
19 (44%)
14 (78%)
48 (63%)
31 (89%)
27 (71%)
14 (82%)
356
203
214
77
138
128
HR* (95%CI)
0.24
(0.11 - 0.54)
0.26
(0.16 - 0.43)
0.78
(0.39 - 1.54)
*HR adjusted with imbalanced covariates
Subgroup
Definition
Size, %
WT
No alteration in PIK3CA AND PTEN AND FGFR1/2 AND CCND1
Single
Single alteration only in PIK3CA OR PTEN OR FGFR1/2 OR CCND1
Two or more alterations in PIK3CA OR PTEN OR FGFR1/2 OR
Multiple
CCND1 genes
Multiple
Minimal
27%
76%
49%
24% 24%
Hortobagyi, SABCS 2013
ESR1 mutation (by cf-DNA) can predict sensitivity to
different hormonal drugs*
*SoFEA cohort
ESR1 mutant
ESR1 wild type
Fribbens, JCO 2016
No difference in effect of palbociclib + fulvestrant
ESR1wt or ESR1mut (by circulating cf-DNA)
*Paloma3 cohort
ESR1 mutant
in
ESR1 wild type
Fribbens, JCO 2016
Challenges for clinician related to
breast cancer heterogeneity
• Understanding heterogeneity
– Looking to the tree or to the wood?
• Measuring heterogeneity
– Dissecting high-throughput data to pick-up the relevant ones
• Monitoring heterogeneity
– Rebiopsy, liquid biopsy, targeted imaging
• Treating heterogeneity
– Patient selection, target’s selection, selective or multitarget drug,
horizontal or vertical multiple blocks, sequence strategies, …
Addressing heterogeneity
in clinical practice
• Decrease heterogeneity by dissecting
tumors by just one homogeneous
characteristic
– unrealistic, due to the number of variations
and to the imprecision of tecniques
Screening breast cancer for actionable
molecular alterations
About 400 MBC pts screeened for genomic alterations
Andre’, Lancet Oncol 2013
Personalised therapy is really possible in a
minority of patients, positive results just in fews.
13% of pts with actionable targets; RR 1%
Andre’, Lancet Oncol 2013
Addressing heterogeneity
in clinical practice
• Decrease heterogeneity by dissecting tumors by just one
homogeneous characteristic (unrealistic, due to the number of
variations and to the imprecision of tecniques)
• Accept heterogenity, and pick-up the master
regulator of the tumor society to induce a
catastrophic crash
– large failure of clinical trials, due to redundance of
escape pathways
Molecular subtyping breast cancer could be useful
to select the right patient for the right drug
Response to primary therapy by intrinsic subtype
in CALGB 40601 neoadjuvant trial
Carey, ASCO 2013
Targeting cell-cycle regulators
in hormone-resistant BC
Turner, NEJM 2015
Addressing heterogeneity
in clinical practice
• Decrease heterogeneity by dissecting tumors by just one
homogeneous characteristic (unrealistic, due to the number of
variations and to the imprecision of tecniques)
• Accept heterogenity, and pick-up the master regulator of the
tumor society to induce a catastrophyc crash (large failure of
clinical trials, due to redundance of escape pathways)
• Accept heterogeneity, and hit multiple targets
simultaneously or in sequence
– many ongoing trials, thougthly marginally improving
the results
Multiple block strategy in neoadjuvant
setting had limited fallout in adjuvant
Different size effect related to hormonal status
Addressing heterogeneity
in clinical practice
• Decrease heterogeneity by dissecting tumors by just one homogeneous
characteristic (unrealistic, due to the number of variations and to the
imprecision of tecniques)
• Accept heterogenity, and pick-up the master regulator of the tumor
society to induce a catastrophyc crash (large failure of clinical trials, due
to redundance of escape pathways)
• Accept heterogeneity, and hit multiple targets simultaneously (many
ongoing trials, thougthly marginally improving the results)
• Limit heterogeneity effects, by monitoring the
changes in tumor composition and the onset of
new master regulator and rapidly adapting
therapies, both by functional imaging and by
molecular analysis
– very difficult for possible divergent behavior of
different metastases
cf-DNA PIK3ca mutations are predictive of efficacy of
BKM120*
* Belle-2 population
Baselga, SABCS 2015
Evolution of heterogeneity
(the case of 60 yr-old BC pt, treated with PIK3CA inhibitor)
variation of allele frequencies (VAF) of the
listed gene mutations in the three lesions
Juric, Nature 2015
Heatmap of the non-silent genetic alterations
across the primary tumour and the 14 metastases
Phylogenetic
evolution of the
metastases, with
bi-allelic loss of
PTEN leading
patient to death
Juric, Nature 2015
Don’t miss host heterogeneity
• At individual level (PK, PD, SNPs, …) and at tissue
level (immune response and inflammation,
angiogenesis, stromal reaction)
pCR by TILs in the CherLOB study (Dieci, Ann Oncol 2016)
Conclusions
• Heterogenity is a natural evolutionary needs of cancer
• Mutation, drift and selection represent the key
mechanisms regulating heterogeneity
• In BC, whole genome analysis revealed high degree of
mutations, insisting at different levels in different cell
pathways.
• Main mutations occur at non coding DNA regions, but
their significance has been not yet elucidated
• Heterogeneity is the limiting-effect of current
therapies, both at a static and a dynamic level
• Better understanding of the development and of the
role of different clones, as well as of their relationships
and fate, is needed to rationally use anticancer drugs