Checkpoint inhibitors Flashcards
Draw diagram of ICIs
Describe T cell activation
Need a number of signals. First is TCR and complementary antigen bound to MHC. If this is only signal = anergy. Second signal is CD28 binding to one of the B7 (CD80 and CD86) on the APC. Cytokine environment then determines which differentiation state the T cell will commit to. To prevent excess activation, T cell upregulates CTLA-4 which also binds B7 (10x higher affinity thna CD28). This is coupled to inhibitory signalling cascade and dampens T cell response
What sort of antigens can tumour express?
can be from oncogenic viruses, differentiation antigens, epigenetic alterations in cancer cells, or neoantigens derived from mutations associated with carcinogenesis
What have we found out in the last 5 years or so about CTLA-4?
- doesn’t just bind to CD80/CD86 on the APC, but it also removes them from the membrane – ‘hoovers them off’
Describe PD-1 mechanism
- Cancer cells under attack from T cells release IFN-y which leads to expression of PD-L1
- PD-L1 binds PD-1 on T cells
- PD-1 phosphorylated on its ITAM motifs in its cytoplasmic tail
- This recruits SHP-2 which keeps the TCR dephosphorylated, preventing TCR signalling
Waterhouse et al. (1995)
- first demonstrated the importance of this regulatory process by creating CTLA4 knockout mice, which consequently had huge and uncontrollable proliferation and activation of T cells, leading to their death.
Leach et al. (1996)
- Was a post-doc working w Jim Allison
- Showed CTLA4 to be a negative regulator of T cell activation, which earned a Nobel Prize.
- They injected BALB/c mice with B7-1 transfected colon cancer cells.
- These mice received an anti-CTLA4 antibody, an anti-CD28 antibody, or remained as untreated controls.
- The anti-CTLA4 antibody led to the rejection of pre-established tumours – they regressed in size or were obliterated completely.
- 90% of the cancers in treated mice disappeared.
- This seminal work resulted in the birth of the first checkpoint inhibitors in cancer immunotherapy – anti-CTLA4 monoclonal antibodies.
Describe anti-CTLA4 mAb approval
Ipilimumab
Approved in 2011 by FDA for metastatic melanoma
Has now doubled 10-year survival for mm
Hodi et al. (2010)
- In a seminal phase III trial
- was shown to significantly prolong the lives of patients with stage 4 metastatic melanoma. The median survival with ipilimumab plus the gp100 vaccine (against tumour antigen) was 10 months compared to 6.4 months with the gp100 vaccine alone. Some cases were even ‘cured’ as they showed durable remission.
- The tail of the curve was a plateau – indicated they were stable – this was v exciting!
- Seems to be permanent!
Robert et al. (2011)
- conducted a phase III trial of over 500 patients with untreated metastatic melanoma, who were assigned to either ipilimumab plus dacarbazine (I + D), or dacarbazine plus placebo. The overall survival of the I + D group was 11.2 months, compared to 9.1 months for dacarbazine alone.
- However, there were significant autoimmune-related side-effects:
- 40% had skin reactions
- some had vitiligo, hepatitis
- N + V
- colitis (which led to diarrhoea).
- This was a new class of side effects that had rarely been seen with conventional therapeutics
- Understanding of why these therapies are linked to such side effects may help autoimmunity in general
- However, there were significant autoimmune-related side-effects:
- Nevertheless, was approved by the FDA for clinical use in 2011 ∴ consensus = its benefits override these SEs
Robert et al. (2015)
- The efficacy of the anti-PD1 antibody Nivolumab was shown in a phase III clinical trial for patients with advanced melanoma (without a BRAF mutation)
- They compared nivolumab (N) treatment to a dacarbazine (D) control group. The 1 year survival rates were ~70% for patients who received N, and less than 40% for patients on D; progression-free survival was 40% for patients on N, and less than 10% for patients on D.
Accordingly, antibodies targeting the PD1-PDL1 axis have been approved as second-line or first-line therapies for an ever-growing list of malignancies, including melanoma, lymphoma, lung cancers, renal cell cancer (RCC), bladder cancer and more.
Whats the difference between sites of action of PD-1 and CTLA-4
- PD1 acts in the TME/peripheral tissues whereas CTLA-4 acts in the lymph nodes
Larkin, 2015
- Since PD1 + CTLA4 regulate different inhibitory pathways on T cells, combination therapy w Abs targeting both molecules were tested - may have additive/synergistic effects
- Phase III clinical trial of nivolumab alone vs nivolumab + ipilimumab, vs ipilimumab alone in pts w untreated metastatic melanom
- Progression free survival = 11.5 months with combination; 6.9 months w N, 2.9 w I
-
BUT highly toxic, more adverse effects in comb group than either alone, but mostly manageable w established treatment guidelines.
- Difference between N alone and combination is not significant to claim comb is better
- But, both or N alone = better than I alone
- This trial used biomarkers → establish PDL1 +ve and –ve tumours but:
- IHC (immunohistochemistry) to look at PDL1 Abs are not that good
- Currently unclear how PDL1 or Ag is processes
- Therefore, difficult to draw confident conclusions of differences in treatment response. More work needed.
-
BUT highly toxic, more adverse effects in comb group than either alone, but mostly manageable w established treatment guidelines.
Example of tumours failing to respond to ICIs
For example, a study in 2018 trialled an anti-PD1 therapy in prostate cancer patients, but in most cases, tumours continued to grow despite therapy. Only ~5% of tumours demonstrated shrinking.
Furthermore, two clinical trials in 2019 showed no differences in PD-1 vs existing standard of care for HCC and also in mesothelioma
What are the key points in the ‘what determines success of ICIs’ section?
- Tumour genomes (TMB and high fitness neoantigens, identifying sensitive subgroups)
- Pt germline genetics
- Immune microenvironment (PD-1 expression, TILs, MDSCs, commensal microbiota)
Define TMB
- total number of somatic mutations per coding area of a tumour genome
Alexandrov et al. (2013)
- global sequencing initiative - analysed the mutations from over 7000 cancer samples from 30 different cancer classes.
- Cancers related to chronic exposure to mutagens e.g. lung cancer (tobacco smoking), malignant melanoma (UV exposure) = highest prevalence of mutations + related to specific mutational signatures
- E.g., malignant melanoma – higher prevalence of ‘signature 7’ = characterised by C>T mutations likely due to formation of pyrimidine dimers (bulky, helix-distorting lesion) resulting from UV exposure
- What’s interesting some of the best initial ICI response rates were observed in carcinogen-driven cancers such as melanoma and NSCLC, which have high TMBs.
- Moreover, the negative studies were typically for cancers with a low TMB.
- Cancers related to chronic exposure to mutagens e.g. lung cancer (tobacco smoking), malignant melanoma (UV exposure) = highest prevalence of mutations + related to specific mutational signatures
Snyder et al. (2014)
- Obtained tumour samples from melanoma pts who were treated w anti-CTLA4.Performed whole-exome sequencing on tumours + matching blood samples (massively parallel seq)
- Degree of clinical benefit was associated w TMB but TMB alone could not predict clinical benefit
- E.g. because some patients with low TMB show good response
- Degree of clinical benefit was associated w TMB but TMB alone could not predict clinical benefit
- Instead, they identified that response can be predicted by high-fitness neoantigens…
- They used genome-wide somatic neoepitope analysis and patient-specific HLA typing identified proposed tumour neoantigens for each patient
- They illustrated a neoantigen landscape that is found specifically in tumours that respond strongly to anti-CTLA4 therapy
- This was validated in a second group of pts
- These predicted neoantigens activated T cells from pts treated w ipilimumab in vivo
- Implications warrants analysing exomes of patients who are being considered for anti-CTLA4 treatment predict response
- It must however be considered that this was a relatively small cohort
- Although the neoepitopes identified may be the most important ones, the in vivo relative contribution of each is unknown
Why is it thought high TMB tumours respond best to ICIs?
- It is thought that high-TMB-tumours respond best to ICI because the more mutations a tumour has, the greater its neoantigen heterogeneity is likely to be tumour will likely have more immunogenic neoantigens (i.e. foreign to the immune system a T cell response is more likely)
Marabelle et al 2020
- Paper in Lancet
- Explored associated between high TMB and outcomes in ten tumour-type specific cohorts from phase 2 KEYNOTE-158 study which looked at anti-PD1 agent in patients with solid tumours
- (NB – did not include breast, prostate and mismatch repair proficient CRC)
- Quantified TMB with NGS sequencing panel
- Although they found high TMB was associated with better overall survival – the difference was only modest
As a result of this, FDA approved pembrolizumab for adults and children with TMB-high solid tumours
- May be a bit too soon to have done this?
- In the study – 10% of pts had treatment-related serious adverse effects and 15% had moderate adverse effects
Gurajao et al (2020)
- Larger cohort - 2500
- They showed that there is a difference between responders and non-responders but this is actually limited to a small number of tumour types (such as lung cancer and melanoma)
- They point out that the relationship between TMB and response is largely due to these tumour types and because the drugs are usually used in these tumour types, there is likely an ascertainment bias (data more likely to include some members of a population than others)
- They also had a look at the cut point that the FDA accepted that defined tumours with high TMB – this was 10 mutations per Mb – they found that according to this cut off, 56% of responders are below this and could be denied effective therapy
- Similarly, pts who have disease with higher TMB than this cut-off also showed not to respond
- Criticism – this is a preprint and has not been subject to peer review
- Difficult, because you would want to do a RCT but these drugs have such good clinical benefit – ethical dilemma
- Leads us to question – could we use ICIs in combination with radiotherapy to increase TMB?
- We have had positive results from clinical trials with ICI or placebo with chemoradiotherapy but we have had no study to date that has a control arm with immunotherapy alone
Le et al 2015
- investigated whether MMR deficiency was related to the response rate and immune-related progression-free survival of two groups of colorectal cancer patients treated with pembrolizumab (anti-PD-1 antibody), one group with MMR deficiency, and the other with an intact MMR system.
- The group also performed whole exome sequencing of each tumour.
- The results were consistent with Snyder’s first description of mutational load and responsiveness to immunotherapy – the MMR deficient tumours contained a mean of 1782 somatic mutations and showed a 78% immune-related progression-free survival, compared to a mean of 73 somatic mutations in MMR proficient tumours, which showed a 0% objective response rate and 11% immune-related progression-free survival.
With the recent approval of nivolumab and pembrolizumab for the treatment of MSI-positive cancers of any histology in 2017, anti-PD1 therapy is the first therapy to be granted FDA approval on the basis of a specific tumour genetic characteristic independent of tumour histology.
MSI (microsatellite instability) results from impaired MMR
Why does MMR deficiency lead to sensitivity to ICIs?
- We now understand that mechanistically, MSI-positive tumours have a high TMB; MMRd generates a high mutational load, particularly through indel mutations, some of which result in frameshifts that produce neoantigens that may be more immunogenic on average owing to their greater sequence divergence from self-peptides.
