Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

CDM > Chemotherapy > Flashcards

Chemotherapy Flashcards

1
Q

Describe different aims of chemotherapy (3)

A
  • as a curative treatment
    • put cancer into complete remission AND it doesn’t come back
  • disease control
    • can’t cure the cancer; reduce its burden for as long as possible
  • palliative treatment
    • use to help with symptoms
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Targets for chemotherapy in cell cycle

A

G1 - growth

S - DNA synthesis

G2 - growth and preparation for mitosis

M - mitosis (cell division)

Tend to target different stages of the cell cycle. Normal healthy cells also affects (i.e. SEs) but malignant cells tend to spend more time dividing therefore more prone to the effects of chemotherapy.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the mechanism of action of alkylating agents?

A
  • cause DNA damage
  • cytotoxic during entire cell cycle
  • alkylating agents cause cross links w/in DNA by adding alkyl groups to guanine bases
  • examples:
    • cyclophosphamide
    • bendamustine
    • chlorambucil
    • melphalan
    • ifosfamide
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the mechanism of action of antimetabolites?

A
  • interferes w/ DNA & RNA synthesis by acting as a substitute for normal bases (C, A, T, G)
  • antimetabolites are cytotoxic durig the S phase of the cell cycle
  • examples
    • purine analogues
      • fludarabine
      • 6-mercaptocupurine (6-MP)
    • pyramide analogues
      • cytarabine
      • gemcitabine
      • azacitadine
    • antifolate
      • methotrexate
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the mechanism of action of anti-tumour antibiotics?

A
  • anthracyclines and others:
    • cytotoxic during the s phase of cell cycle
    • intercalate between base pairs
    • inhibit topoisomerase II
    • create oxygen free radicals
  • examples include:
    • daunorubicin
    • doxorubicin
    • epirubicin
    • idarubicin
    • bleomycin
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the mechanism of action of mitotic inhibitors/plant alkaloids?

A
  • vinca alkaloids inhibit microtubule polymerisation
  • mitotic inhibitors interfere with the mitosis phase of the cell cycle
  • examples:
    • vincristine
    • vinblastine
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the role of steroids in chemotherapy?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

List the generic symptoms of chemotherapy:

A
  • myelosuppression
    • cytopenias (pancytopenia)
      • neutropenic fever & neutropenic sepsis
  • gut toxicity
    • sore mouth, change in taste
    • diarrhoea
    • neutropenic colitis/typhylitis
  • N&V
  • hair loss
  • effect on fertility
  • liver toxicity
  • teratogenicity
  • fatigue
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Specific side effects of anthracyclines

A
  • cardiac toxicity
    • seems to be idiopathic –> cannot predict who it will effect
    • baseline echo before starting
    • can cause arrhythmias or affect ejection fraction
  • ‘lifetime dose’
    • due to cardiac toxicity cannot exceed certain dose over lifetime
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Side effects of vinca alkaloids

A
  • peripheral neuropathy
  • constipation
    • can be problematic for older patients
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Side effects of bleomycin

A
  • interstitial lung damage/fibrosis
    • avoid in older people/smokers/chest Hx
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Side effects of ifosfamide

A
  • encephalopathy
  • usually in-patient
  • may need to be halted/give antidote - encephalopathic Sx usually reversible once chemotherapy stops
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Long-term vs short-term side effects

A

Short-term: generic SEs tend to be short-term

Long-term:

  • cardiotoxicity, interstitial lung damage
  • fertility problems
  • bone problems (particularly high dose steroids)
  • secondary malignancies
    • include haematological cancers
      • makes healthy cells more susceptible to developing malignancies in the future
      • e.g. MDS, AML
    • skin cancers
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

How do monoclonal antibodies (MOABS) work?

A
  • works by targeting cancer cell protein and inducing antibody-dependent cellular cytotoxicity and complement dependent cytotoxicity
  • rituximab
    • binds to cell surface CD20 protein
    • CD20 expressed by B cells
  • obinutuzumab
    • 2nd generation anti CD20 monoclonal antibody
  • daratumumab
    • anti CD38 monoclonal antibody
  • SEs:
    • infusion related reactions
      • fever
      • hypotension
      • rash
      • anaphylaxis
    • increased susceptibility
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Mechanism of action of targeted therapies:

A

Target specific cell pathways/proteins/receptors to inhibit cancer cell activation/proliferation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are ibrutinib/acalabrutinib used for?

A
  • B cell receptor inhibitor, Bruton tyrosine kinase inhibitor
  • chronic lymphocytic leukaemia, mantle cell lymphoma
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What are imatinib/dasatinib/nilotinib/ponatinib used for?

A
  • tyrosine kinase inhibitor
  • chronic myeloid leukaemia, Ph positive acute lymphoblastic leukaemia
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is venetoclax used for?

A
  • Bc12 inhibitor
  • chronic lymphocytic leukaemia, acute myeloid leukaemia
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is midostaurin used for?

A
  • FLT3i
  • acute myeloid leukaemia
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What are lenalidomide/thalidomide used for?

A
  • immunomodulatory agents
  • myeloma
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is a bone marrow transplant?

A
  • Bone marrow transplant = human stem cell transplant
  • Note: we rarely use true bone marrow now as bone marrow stem cells are collected peripherally but the term bone marrow transplant is still used
  • what is being transplanted is an immune system
  • takes place w/ conditioning = chemotherapy +/- radiotherapy
  • then replacing stem cells
  • sources:
    • unrelated –> transplant registry
    • rarely frm umbilical cord
    • other realtives –> haploid (half matched) - sibling, parent, child
      • increases pool of potential donors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Why do bone marrow transplants?

A
  • Haem malignancies
    • most common by far in adults
    • most commonly AML/ALL
    • other malignant bone marrow conditions too e.g. lymphoma, myelodysplasia
  • red cell disorders
    • e.g. sickle cell disease
  • immunodeficiencies
    • SCIDs
  • others:
    • especially in paediatric practice such as metabolic diseases
24
Q

What is involved in bone marrow transplants?

A
  • identify a donor/harvest stem cells
    • autologous: harvest and freeze
    • allogenic: harvested then given to patient, fresh usually
  • hospital admission
  • chemo +/- radiotherapy
  • stem cell return
    • like blood transfusion, through an indwelling line, usually a Hickmann line
  • await engraftment
    • cells overtime move to bone marrow
    • remain in hospital during this time as most vulnerable to infection
25
Q

Major risks and side effects of HSCT?

A
  • side effects as per chemotherapy
    • nausea, vomiting, hair loss, infertility, organ toxicity
  • infection
  • graft vs host disease - allogeneric only not autologous
  • relapse of original disease
26
Q

Graft versus host disease

A
  • unique to allogeneric transplant
  • ‘opposite of rejection’
  • donor immune system attacing recipient
  • HLA matching - not really a ‘perfect match’
  • acute vs chronic
    • acute tends to be in the first 100 days
    • chronic after, but not sstrictly time defined - there is some overlap between the two of them
    • defined more by symptoms
27
Q

Acute GVHD

A
  • skin - rash, can be anywhere on the body
    • erythematous, macular papular, often very itchy
  • GI tract - nausea/diarrhoea
    • often +++ litres of watery diarrhoea
  • liver
    • tends to be reflected on blood tests opposed to Sx
28
Q

Chronic GVHD

A
  • major cause of late morbidity and non-relapse post HSCT
  • clinical features much more varied than acute, often resemble autoimmune diseases
    • joint stiffness/swelling
    • dry eyes
    • dry mouth
  • often develops post-acute GVHD
29
Q

GVHD - effect on outcome

A
  • poeple with GVHD are less likely to relapse
  • they have significantly better long-term survival than the patients who did not develop any GVHD
  • therefore we don’t give loads of immunosuppression to avoid GVHD
30
Q

How effective is HSCT?

A
  • 30% risk of disease relapse despite transplant
  • 10-20% transplant related mortality
31
Q

PD-1 inhibition as treatment for cancers

A

Cancer cells use PD-1 upregulation as a way to ‘cloak’ from T-cells

  • very good results in patients with Hodgkin disease
  • used extensively in solid cancers
  • (did not work in other haem cancers)

BUT

  • significant side effects especially autoimmune disorders
    • PD-1 is down-regulated on normal cells too
    • can be life-threatening
    • can affect any organ
32
Q

CAR-T cells

A

Works by transfusing own T-cells with a vector which contains a new T-cell receptor which recognises any antigen you want to target. It takes about 4-5/52 to generate the CAR-T cells. They can be infused to the patient after they have received some chemotherapy.

33
Q

Bispecific antibody

A

Antibody which links T-cells via the CD3 receptor with your target of interest via specific antibodies. Brings T-cells into close proximity to the carcinogenic cells –> cell killing.

34
Q

Outcomes of CAR-T cells in refractory DLBCL (historical vs ZUMA-1)

A
  • ZUMA-1 study: patients with relapse diffuse large B cell lymphoma were treated w/ CAR-T cells
    • progression free survival significantly better compared to historical controls taken from the SCHOLAR-1 study –> drug approved by FDA & NICE, currently available in the UK
35
Q

Tisagenlecleucel in children and TYA with R/R ALL

A

Tisagenlecleucel = another CAR-T cell product

  • used in children and young adults with multiple relapse ALL
    • typically carries v poor porgnosis

With this treatment:

  • 12 months event free survival 50%
  • 12 months overall survival 76%
  • prognosis significantly improved –> NICE approved the drug
36
Q

Cytokine Release Syndrome (CRS)

A

Systemic inflammatory response cause by cytokines released by CAR-T cells and other immune cells and results in reversible organ dysfunction. CAR-T cell therapy has acute toxicity due to the release of cytokines when the T-cells get activated.

During 1st 10-14/7 high chance that pts will develop CRS. Can affect every organ however typically present with:

  • drop in BP
  • SOB
  • high temp
  • beginning confusion

Many patients need to be admitted to ICU and so treatment has to be given as an inpatient.

37
Q
A
38
Q

Machine learning - supervised

A
  • goal of supervised learning is prediction of a known outcome
    • i.e. predict outcome based on a labelled training set
    • e.g. predict survival following allogeneic haematopoietic stem cell transplantation
39
Q

Machine learning - unsupervised

A
  • not about predicting a specific output, the algorithm instead attempts to identify patterns/groupings w/in the data
  • e.g. identify clusters of similar gene-expression profiles among B-cell lymphoma samples
40
Q

Machine learning - reinforcement

A
  • newer class of learning - hybrid of supervised & unsupervised learning
  • algorithm maximises accuracy by trial and error
    • goal: optimise prediciton by trial and error
  • E.G. propose optimal vasopressor administration for critically ill patients based on feedback from previous experience
41
Q
A
42
Q

Define ‘problem understanding’

A
  • basically the hypothesis
  • needs to be clear w/ specific aims & type of learning defined to be clinically useful
43
Q

Define ‘data understanding’

A
  • properties are examined using desciptive statistics & visualisation
  • data quality control = key
44
Q
A
45
Q

Define ‘data preparation’

A
  • pre-processing = preparing the dataset for analysis so that it could serve as input for the algorithm
  • things that need to be accounted:
    • data structure & heterogenicity
    • frequency of sampling
    • missing values
    • inclusion of omics
46
Q

Define ‘data modelling’

A
  • decision on which machine-learning algorithm to use
    • depends on type of data
    • depends on goals of prediction
      • broadly divided: regression & classification
        • regression = prediction of continuous properties (e.g. length of hospitalisation)
        • classification = discrete events (e.g. development of neutropenic fever yes/no)
47
Q

Define ‘evaluation’

A
  • fundamental measures of model performance are its discrimination and calibration
    • discrimination = most often quantified by C-index (range 0.5-1)
      • reflects probability that for any randomly elected pait of individuals, one w/ and one w/o the outcome, the model assigns a higher probability to the individual with the outcome
      • perfect discrimination = C-index 1 = predicted risks for all individuals who have (diagnostic) or develop (prognosis) the outcome are higher than those for all individuals who do not experience th outcome
    • calibration = agreement between observed end-points & predictions
      • e.g. if we predict a 10% risk that a pt will die w/in 1 year, the observed outcome in the test set should approximate 10 deaths per 100 pts
  • depending on model’s application, these measures can also be reported:
    • interpretability
    • sensitivity
    • specificity
    • +ve/-ve predictive values
48
Q

Define ‘deployment’

A
  • once it has been built & validated, can be released for implementation
  • optimally would be integrated into clinical eorkflow w/ ongoing effort to assess effectiveness
    • measures include
      • volume of users
      • feedback from users
      • degree of clinical reliance on the system
  • should ideally also undergo real-time evaluation & be updated & re-validated accordingly
49
Q

Limitations of machine learning - data availability

A
  • machine learning is frequently used in high-dimensional settings
  • often ‘data-hungry’
    • requiring extensive databases before providing useful results
  • need for manual input of individual data sets
  • changes in electronic medical records may eventually overcome this limitation
50
Q

Limitations of machine learning - bias and data quality

A
  • often relies on unstructured data sources that were designed to serve other purposes
  • equality of data introduced onto the model varies
    • some are closely audited, some are loosely monitored
  • erroneous learning can result from inconsistent classification of features & outcomes included in models across practitioners & centres
  • identification of potential biases at the study design phase as well as testing the model at variety of populations = measures to reduce bias
51
Q

Limitations of machine learning - interpretability

A
  • ‘black box models’
    • machine-learning algorithms generate models w/o providing insight into the logic between the association between predictors & outcomes
    • runs counter to the thought process of clinicians
    • however, note that medical problems are often high-dimensional & there are cases where the logic is beyond perception
52
Q

Limitations of machine learning - misapplication

A
  • advantage of machine learning is mainly in the high-dimensional setting
    • where standard regression methods may be overwhelmed
    • (machine learning models are percieved by many to outperform conventional methods for prediction)
  • in settings where the number of predictors is not high, there is no inherent benefit to machine learning
  • some degree of randomness in a pt journey = not all events are predictable
53
Q

Limitations of machine learning - the curse of dimensionality

A
  • may be adventageous in scenarios where the number of features is high relative to the number of cases
  • issues arising when working w/ high-dimensional datae
  • using feature selection techniques, adhering to the standards of prediction model development and external validation and including clinical reasoning may mitigate some of these problems
54
Q
A
55
Q

Limitations of machine learning - ethics

A
  • ‘dataism’
    • dependency of humans on algorithms and data over judgement
  • at the final decision point, clinical judgement of experienced physicians remains paramount
  • patient privacy at risk w/ integration of large data streams

CDM (7 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions