The invention relates to an isolated T-cell population most suitably for use in T-cell therapy, characterized in that said T-cell population has a mean telomere length, measured using high-resolution telomere length analysis, that is greater than or equal to a threshold value; and methods for establishing the maximum number of populations doublings of an isolated T-cell population before the T-cells become functionally ineffective; methods for assessing the suitability of an isolated T-cell population for use in T-cell therapy; T-cell therapy using said isolated T-cell population; and methods for isolating a population of T-cells from a sample.

Background of Invention

T-cell therapy, also known as adoptive transfer of T-cells, including T-cells engineered to express a chimeric antigen receptor (CAR), or antigen-selected tumour infiltrating T-cells, has emerged as a very promising type of targeted immunotherapy. Some studies have shown complete remission rates as high as 95% in relapsed and refractory acute lymphoblastic leukaemia (ALL) ¹ . These new drugs are categorized as adoptive cellular therapies; the two agents approved thus far utilise the patients’ autologous T-cells and re- engineer them to target cancer cells expressing CD19 for the treatment of B- cell leukaemias and lymphomas.

Study results show certain patients respond well, particularly compared with conventional treatment options, but because experience with these novel therapies is still limited, there are questions about response durability. It is of clear importance to be able to select out patients who will respond well to adoptive cellular therapies, including CAR-T cell therapy, given that only 26% of those with relapsed/refractory Chronic lymphocytic leukaemia (CLL) have previously demonstrated durable anti-tumour complete responses in clinical trials ² . Due, in some measure, to the cost of these personalised medicines (currently up to $475,000/ patient), the concept of allogeneic (“off the shelf) CAR-T cells has spawned much interest. So far, clinical trials using donor-derived CD19 CAR-T cells in patients with relapsed B cell malignancies have shown potent anti-tumour responses without substantial alloreactivity ³ ’ ⁴ However, question marks remain over the duration of response to these agents. Therefore, it remains a major objective to identify biomarkers that can predict both response and duration of response to these cell-based therapeutics.

Telomeres are structures that cap the ends of eukaryotic chromosomes preventing their recognition and repair as a double stranded DNA break. As a consequence of the inability of semi-conservative replication to fully replicate linear DNA molecules, telomeres are subjected to gradual erosion with ongoing cell division ⁵ . Ultimately, the loss of the capping function triggers a p53 dependent G1/S cell cycle arrest that blocks further cell division, referred to as replicative senescence ⁶ . Thus, the replicative lifespan of many human cell types is governed by telomere length. The telomere length at which replicative senescence is induced in human cells has not been functionally established. However high resolution, single telomere length analysis (STELA) has provided the lower length ranges of telomeres detected in senescent cells at the limit of their replicative lifespan. Using this technology mean telomere lengths in senescent fibroblast cells have been detected at 2.0 kb, with single alleles displaying mean telomere lengths as low as 1 .0 kb ⁷ . Telomere length can therefore be utilised as a measure of replicative history and capacity, providing a marker of biological age.

The majority of adoptive T-cell products are autologous, being manufactured from individual patient-derived T-cells. However, there is also interest in developing allogeneic T-cell products, that have the potential to provide‘off the shelf products for a broad range of patients. Patient (or donor)-derived T-cells have a limited replicative capacity that is governed by their telomere length. This can provide a natural limit to the long-term anti-tumor effects of CAR-T cells, and other adoptive cell transfer approaches, which require an extensive replicative capacity to allow for both ex vivo expansion and then ongoing proliferation in vivo in order to sustain a therapeutic response. Therefore, as cell-based therapeutics become mainstream, it will be increasingly important to optimise the manufacturing process to ensure the best possible personalized product is available to enable the maximal therapeutic response. Importantly also, it will be necessary to identify those patients who would particularly benefit from these treatment modalities and those who may not due to the biological age, and/or the adverse characteristics, of their T-ceil pool.

Many human cancers are associated with aberrant immune cell ageing. This often manifests itself as skewing towards memory phenotypes in the T-cell pool and a concomitant reduction in naive cells. Memory ceils typically have shorter telomere length profiles and although these cells are capable of further expansion, their replicative capacity is diminished. These cells also have a reduced capacity to migrate into tissues. This may be critical in determining the long-term efficacy of T-celi therapies, including CAR-T; T-cell exhaustion and/or senescence will negatively impact on T-celi (therapeutic) function. Moreover, the heterogeneous replicative history of T-cell subsets has the potential, particularly in patients with signs of immune cell ageing, to further skew the T-cell repertoire when T-cell populations are driven to expand ex vivo.

To address the problem of identifying a biomarker that can accurately predict both response and duration of response in T-cell based therapeutics, this study established the relationship between telomere length and T-cell functionality (specifically the secretion of pro-inflammatory cytokines such as, without limitation, interleukin-2 (IL-2)). Counterintuitively, it was observed that the loss of T-cell function occurs at a longer telomere length threshold than that detected in replicative senescent cells. Consistent with this, T-cells that had lost function at a defined telomere length, were still capable of cell division following stimulation. This telomere-length threshold for T-cell function has important implications for defining therapeutic response and patient selection in adaptive immune cell therapies.

In our investigations, we have used telomere length analysis to identify a telomere length threshold that may be used as a biological parameter that is highly prognostic for response to treatment. Furthermore, given the known relationship between telomere erosion and cell division in T-cells, this threshold value can be modified and/or raised depending on the number of population doublings required to generate any required expansion of T-cell based therapeutics

Statements of Invention

According to a first aspect of the invention, there is provided an isolated T-cell population for use in T-cell therapy, wherein said T-cell population has a mean telomere length, measured using high-resolution telomere length analysis, that is greater than or equal to a threshold value, wherein said threshold value is at least 3.47 kb.

Most preferably said threshold is at least between 3.47 and 4.34 kb.

By ensuring that any T-cell based therapeutic has a telomere length above this threshold value, said T-cells not only retain their replicative capacity, but also retain a sustained functional response (i.e. the ability to secrete one or more pro-inflammatory cytokines, such as those shown to that sustain T-cells and even CLL B-cells).

In addition, given that there is a well-defined relationship between telomere erosion and cell division in T-cells, with telomere erosion rates varying between 107-120 bp per population doubling ^8-10 , the threshold value can be raised above the minimum 3.47 kb threshold, depending on the number of population doublings to be undertaken during a T-cell therapy program. In such cases, the threshold value is 107 bp - 120 bp multiples greater than 3.47 kb. More preferably, the threshold value is 110 bp multiples greater than 3.47 kb. In particularly preferred embodiments, the threshold value is calculated according to the following formula

Threshold value = 3.47 kb + (X) _n

wherein:

X ranges from 0.107 kb to 0.120 kb, and is more preferably 0.11 kb; and n is an integer between 1 and 50 and represents at least the minimum number of population doublings said T-cell population is to undergo during said T-cell therapy.

In particularly preferred embodiments of the first aspect of the invention, the isolated T-cell population has a mean telomere length that is greater than or equal to 3.47 kb and less than or equal to 10 kb, ideally 9.47 kb.

In some embodiments of the first aspect of the invention, the isolated T-cell population of the first aspect of the invention is for use in ex vivo amplification and, ideally, multiple rounds of ex vivo amplification.

Preferably, the isolated T-cell population is for use in a T-cell therapy program, adoptive T-cell therapy or CAR-T therapy. Additionally or alternatively, the isolated T-cell population may be used to carry genetically modified T-cell receptors, may be genetically modified T-cells (i.e. modified outside of the TCR locus), neoantigen reactive T-cells, tumour infiltrating lymphocytes or CAR Natural Killer cells

In such embodiments, the T-cell population may or may not be subjected to ex vivo amplification and typically multiple rounds of ex vivo amplification as part of said program.

In preferred embodiments, the isolated T-cell population are autologous and, in such embodiments, said T-cell population is preferably obtained from a patient presenting with a cancerous condition.

In a preferred embodiment said cancer is a B-cell malignant condition. B-cell malignant conditions include, but are not limited to, B cell Acute lymphoblastic leukaemia (B-ALL), Diffuse Large B-cell lymphoma (DLBCL), Non-Hodgkin Lymphoma (NHL), Mantle Cell Lymphoma (MCL), Hodgkin Lymphoma (HL) and Multiple myeloma (MM). Preferably, the B-cell malignant condition is chronic lymphocytic leukaemia (CLL).

In alternative embodiments of the invention said T-cells are allogeneic.

High resolution telomere length analysis is typically measured using a single telomere length analysis (STELA) method, which allows complete resolution of telomere lengths at specific chromosome ends ⁷ ’ ¹¹ . However, any other method that can measure the full range of telomere length from one TTAGGG repeat to several kb of telomere length may be utilised. In some embodiments, telomere length is detected for a single chromosome. Alternatively, a mean telomere length may be detected for a plurality of different chromosomes and, in such embodiments, an average mean telomere length is calculated. In preferred embodiments, a mean cell population telomere length is determined for the XpYp chromosome, but yet more preferably a mean telomere length is determined for a plurality of different chromosomes. Preferably, these chromosomes include XpYp, 16p, 17p, 2p, 18q, 11 q, 9p, 7q, 5p and/or 12q, and most preferably these chromososomes include XpYp, 17p and/or 7q.

By careful selection of the telomere length threshold value, the T-cell population has been shown not only to be capable of replication and expansion, but is also functionally effective. By functionally effective we mean, said T-cells retain the ability to secrete one or more pro-inflammatory cytokines, which, ideally, promote the survival and/or growth of T-cells.

In yet a preferred embodiment said T-cells are said to be functional when they secrete cytokines that promote the survival and/or growth of B-cells, such as CLL B-cells.

Pro-inflammatory cytokines include but are not limited to interleukin-2 (IL-2), interferon gamma (INFy), interleukin-6 (IL-6) and tumour necrosis factor alpha (TNFa). In particular, the functionally effective T-cells retain the ability to secrete IL-2, which has been shown to strongly correlate with sustaining cell growth e.g. CLL B-cells growth.

In preferred embodiments of the first aspect of the invention, all cells within the isolated T-cell population have a minimum telomere length, measured using high-resolution telomere length analysis, that is greater than or equal to a minimum value, wherein said minimum value is at least 3.47 kb, preferably at least 3.58 kb and still more preferably at least 3.69 kb.

According to a second aspect of the invention, there is provided a method for establishing the maximum number of population doublings of an isolated T-cell population before the T-cells become functionally ineffective, the method comprising:

a. using high-resolution telomere length analysis to determine the mean telomere length (in kb) of at least a sample of cells from said isolated T-cell population; and

b. calculating said maximum number of population doublings according to the following formula:

mean telomere length 3.47

pop. doublings - -

X

wherein X ranges from 0.107 to 0.120, and is preferably 0.11.

According to a third aspect of the invention, there is provided a method for assessing the suitability of an isolated T-cell population for use in T-cell therapy, the method comprising using high-resolution telomere length analysis to determine the mean telomere length (in kb) of at least a sample of cells from said isolated T-cell population; and

i. concluding that the T-cells are suitable for use in T-cell therapy when said mean telomere length is greater than or equal to a threshold value; or ii. concluding that the T-cells are unsuitable for use in T-cell therapy when said mean telomere length is less than said threshold value; wherein said threshold value is at least 3.47 kb. According to preferred embodiments of the third aspect of the invention, the threshold value is calculated according to the following formula:

Threshold = 3.47 kb + (X) _n

wherein X ranges from 0.107 kb to 0.120 kb and is preferably 0.1 1 kb; and wherein n is an integer between 1 and 50 and represents at least the minimum number of population doublings said T-cell population is to undergo during said T-cell therapy.

According to a fourth aspect of the invention, there is provided a method of T- cell therapy, the method comprising: i. using high-resolution telomere length analysis to determine the mean telomere length (in kb) of at least a sample of cells from an isolated T-cell population; and ii. where said mean telomere length is greater than or equal to a threshold value, concluding that the T-cells are suitable for use in T- cell therapy; and iii. infusing said isolated T-cell population into a patient; but iv. where said mean telomere length is less than said threshold value; v. rejecting said population of said cells; wherein said threshold value is at least 3.47 kb.

Like the third aspect, according to preferred embodiments of the fourth aspect of the invention, the threshold value is calculated according to the following formula:

Threshold = 3.47 kb + (X) _n wherein X ranges from 0.107 kb to 0.120 kb and is preferably 0.1 1 kb; and wherein n is an integer between 1 and 50 and represents at least the minimum number of population doublings said T-cell population is to undergo during said T-cell therapy.

According to a fifth aspect of the invention, there is provided a method for isolating a population of T-cells from a sample comprising: a. using high-resolution telomere length analysis to determine the mean telomere length (in kb) of at least a sample of cells from said population of T-cells; and

b. where said mean telomere length is greater than or equal to a threshold value of at least 3.47 kb;

c. extracting said T-cells from said sample.

In some embodiments of any, some or all of the second to fifth aspects of the invention, high-resolution telomere length analysis is undertaken after any ex vivo amplification step and prior to infusion of said isolated T-cell population into a patient. In alternative embodiments, high-resolution telomere length analysis is undertaken prior to subjecting the isolated T-cell population to ex vivo amplification and typically multiple rounds of ex vivo amplification.

In preferred embodiments of any, some or all of the second to fifth aspects of the invention, the T-cell population is autologous and is obtained from a patient, ideally one presenting with a malignant condition. In particularly preferred embodiments, the malignant condition is a B-cell malignancy such as chronic lymphocytic leukemia (CLL).

As is the case for the first aspect of the invention, mean telomere length may be determined in any or all of the second to fifth aspects of the invention for a single chromosome. Alternatively, mean telomere length may be detected for a plurality of different chromosomes and, in such embodiments, their average mean telomere length is used in the above methods as the threshold value. In preferred embodiments, mean telomere length is determined for the XpYp chromosome, either alone or in combination with one or more further chromosomes such as one or more of 17p, 2p, 16p 18q, 11 q, 9p, 7q, 5p and 12q, and preferably in combination with one or more of 17p and 7q.

In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word“comprises”, or variations such as“comprises” or“comprising” is used in an inclusive sense i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

All references, including any patent or patent application, cited in this specification are hereby incorporated by reference. No admission is made that any reference constitutes prior art. Further, no admission is made that any of the prior art constitutes part of the common general knowledge in the art.

Preferred features of each aspect of the invention may be as described in connection with any of the other aspects.

Other features of the present invention will become apparent from the following examples. Generally speaking, the invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including the accompanying claims and drawings). Thus, features, integers, characteristics, compounds or chemical moieties described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein, unless incompatible therewith.

Moreover, unless stated otherwise, any feature disclosed herein may be replaced by an alternative feature serving the same or a similar purpose. The invention will now be described by way of example only with reference to the following figures: Figure 1. Correlation of telomere length with age in different lymphocyte subsets.

(A) Normal B-cell telomere length are strongly associated with the age of the donor. In contrast, (B) CLL B-cell and (C) T-cells derived from CLL patients’ telomere length are not associated with age. The lower detectable telomere length threshold is shown by the dotted red line (D) Telomere length, determined using STELA, in normal human peripheral blood samples. The linear regression line (solid red), 95% Cl (dotted red) and the lower detectable telomere length threshold (dotted blue line) are shown.

Figure 2. Correlation between telomere length and duration of disease.

(A) There was only a weak correlation between CLL B-cell telomere length and autologous T-cell telomere length. (B) T-cell telomere length was strongly associated with time since diagnosis. (C) demonstrates on-going T-cell telomere erosion throughout the course of the disease whereas (D) CLL B- cell telomere length is not significantly associated with the time since diagnosis. (E) The strong correlation between T-cell telomere length and time since diagnosis was also shown following analysis of a larger cohort of CLL patient samples, whereas no significant correlation between B-cell telomere length and time since diagnosis for this larger cohort of CLL patient samples. Figure 3. Ki-67 expression in CD3 ⁺ T-cells and CLL B-cells in co-culture.

(A) Autologous T-cells showed a rapid increase in Ki-67 expression during co- culture. (B) The expression of Ki-67 in autologous T-cells correlated with the viable cell count in the cultures, which were predominantly CLL B-cells (>95%). (C) CD3 depletion of the cultures significantly reduced the expression of Ki-67 in CLL B-cells and was associated with (D) early termination of the cultures due to loss of CLL cell viability. Figure 4. Autologous T-cells produce cytokines that sustain CLL B-cells in vitro.

(A) The supernatants from the co-cultures showed elevated levels of IL-2, IL- 6, INFy and TNFa. (B) Autologous T-cells were responsible for the production of these cytokines as CD3-depleted cultures showed significantly reduced levels of all four cytokines. (C) The levels of IL-2 in the culture strongly correlated with the percentage of (C) CD3 ⁺ T-cells expressing Ki-67 and (D) CLL B-cells expressing Ki-67.

Figure 5. CFSE labelling of T-cells derived from co-culture that had stopped showing signs of growth.

T-cells were isolated from co-cultures by negative selection (CD19 beads) and were then labelled with CFSE prior to stimulation with CD3/CD28 beads for 3 days. The overlaid histograms show that Although the T-cells in the culture has stopped secreting pro-inflammatory cytokines and showed very low expression of Ki-67, they were still capable of further cell division following stimulation.

Figure 6. Autologous T-cells from patients with CLL show telomere shortening during the disease.

(A) Serial T-cell samples derived from patients with CLL demonstrate significant telomere shortening over the course of their disease. This was associated with increased clonality of the T-cell repertoire as evidenced by (B) a reduction in the standard deviation of for the telomere length distributions and (C) an increase in the abundance of the most frequent T-cell clones over time. (D) T-cell telomere length at the start of the in vitro co-culture determines the duration of proliferation and survival of both the T-cells and the leukaemic B-cells. METHODS

Purification of T-cells and tumour B-cells

Dynabead separation

CD3 ⁺ T-cell and CD19 ⁺ B-cells were positively selected by using CD3 ⁺ and CD19 ⁺ dynabeads from Life Technologies. 10pL of the desired bead was used for the isolation of 500,000 target cells form isolated mononuclear cells. The cell preparation containing the cells of interest and the washed dynabeads was incubated for 20 minutes at 4°C. Cells were isolated by placing the microtube back into the magnetic particle concentrator for 2 minutes. After this time, the supernatant was aspirated and discarded and the microtube was removed from the magnetic particle concentrator.

FACS separation

For small numbers of target cells (less than 100,000), cells were isolated following the staining of samples with fluorescence-labelled antibodies recognising CD19-APC (Allophycocyanin), CD5-FITC (fluorescein isothiocyanate) and CD3-PE (phycoerythrin) and a BD FACS Aria (BD Biosciences). Purity checks of the isolated target cells were performed and, in every case, >97% purity was achieved.

Telomere length analysis using STELA

DNA was extracted using standard proteinase K, RNase A, phenol/chloroform protocols or the QIAamp DNA Blood Mini Kit (Qiagen, Manchester, UK). For telomere length analysis at the XpYp telomere we used the single telomere length analysis (STELA) assay, as previously described ⁷ ’ ¹¹ . CD40L co-culture

CD40L-expressing mouse fibroblast cells were suspended in 5ml_ of fibroblast media and irradiated (8000RADs). 1 x10 ⁶ irradiated cells were then seeded into a 6-well plate in 3m L of fibroblast media. The plate was incubated for 24 hours at 37°C with 5% CO2 to allow the fibroblast cells to adhere to the well surface. Subsequently, duplicate aliquots of 20x10 ⁶ CLL patient’s PBMCs (Peripheral blood mononuclear cells) were taken and re-suspended in 3ml_ of primary culture medium - Roswell Park Memorial Institute (RPMI) 1640 (Sigma) was supplemented with 2mM L-Glutamine (Life Technologies), 100units/ml Penicillin and 100g/mL streptomycin (Life Technologies) and 10% FCS (Life Technologies)) supplemented with 5ng/pL IL-4 (Biosource). 4mM fludarabine (Fludarabine purchased from Bayer Pharmaceuticals and stored as a stock 100mM solution at -20°C) was added to one of these samples, whist the other was unmodified as the control. These samples were left to incubate in 15ml falcon tubes at 37°C in a 5% CO2 moist chamber for 1 hour. The 20x10 ⁶ CLL patient’s PBMCs were then added to the surface of the irradiated cells and incubated at 37°C in a 5% CO2 humidified chamber. On day 3 or 4, 1 mL of fresh culture medium was added to each sample well. Every 7 days the cells were transferred to new CD40L feeder wells.

Ki-67 expression and apoptosis assessment by flow cytometry

Ki-67 expression

Intracellular Ki-67 staining was used to identify cells in active phases of cell cycle. Cells were taken from the co-cultures, washed once in 1 mL of PBS and surface stained with antibodies recognising CD19-APC and CD3-PE. Cells were incubated at 4°C for 15 minutes before being washed twice in 1 mL of

PBS. 60pl of Reagent A (Fixation medium, Life Technologies) was added and the cells were incubated again for 15 minutes at 4°C to fix the cells. Following this, cells were again washed twice in 1 ml of PBS before 56pL of reagent B (Permeabilisation Medium, Life Technologies) supplemented with 10 %v/v NP40 (10g tergitol-type NP40 (nonyl phenoxypolyethoxylethanol) (Sigma) prior to the addition of 5pL of a Ki-67-FITC antibody (Alere). The cells were then incubated for a further 15 minutes before finally being washed twice more with 1 ml_ of PBS and re-suspended in a final volume of 200pl_ of PBS for flow cytometric analysis.

Apoptosis

Cells were then harvested and then resuspended in 195mI of a calcium-rich buffer. Cells were then stained with APC-conjugated CD19, PE-conjugated CD3 and FITC-conjugated Annexin V. Using an Accuri C6 flow cytometer, a gating strategy was employed to quantify apoptosis in CD19 ⁺ B-cells and CD3 ⁺ T-cells, with appropriate compensation applied.

Cytokine assay

1 ml_ aliquots of the supernatants from the long-term culture were harvested and the stored at -20°C following centrifugation to remove cellular material (300 xg for 5 minutes). Subsequently, the supernatant levels of IFN-g, IL-2, IL- 6, IL-10, and TNFa were measured using the V-plex Proinflammatory Panel 1 (MSD). All assays were performed according to the manufacturer’s instructions. In all cases, the supernatants were analysed without dilution and were run in duplicate. The data were acquired on the V-plex® Sector Imager 2400 plate reader and analysed using the Discovery Workbench 3.0 software (MSD). The standard curves for each cytokine were generated using the premixed lyophilised standards provided in the kits. Serial 4-fold dilutions of the standards were run to generate a 7-standard concentration set, and the diluent alone was used as a blank. The cytokine concentrations were determined from the standard curve using a 4-parameter logistic fit curve to transform the mean light intensities into concentrations. The lower limit of quantification (LLOQ) was determined for each cytokine and for each plate as the lowest standard for which both duplicate measurements were above the duplicate background values. The upper limit of quantification (ULOQ) was determined for each cytokine and for each plate as the highest standard that did not reach saturation. Kruskal-Wallis one-way analysis of variance (ANOVA) followed by Dunn’s post hoc test was used to determine the significance, and Mann-Whitney U test was used to determine the p-values between the comparators; a p<0.05 was considered to be statistically significant.

CFSE labelling

Purified CD3 ⁺ T-cells (negative selection), derived from long-term cultures which had stopped showing signs of growth, were labelled with Carboxyfluorescein succinimidyl ester (CFSE) at a final concentration of 2.5mM (Biolegend). The cells were then stimulated with human T-activator CD3/CD28 beads (Life Technologies) plus 30 U/mL of recombinant IL-2 (Miltenyi) for 3 days. CFSE fluorescence was then assessed using flow cytometry and compared with unstimulated cells.

RESULTS

CLL B-cell and matched T-cell telomere lengths from CLL patients are not associated with age

We have observed that normal B-cells (CD19VCD5 ) derived from CLL patients show a statistically significant inverse correlation between age and mean telomere length (r ² =-0.61 ; P=0.01 , Figure 1A). In contrast to the normal B-cell subsets, there was no correlation between malignant CLL B-cell telomere length and age in the patient cohort (r ² =0.01 , P=0.30; Figure 1 B), suggesting that the telomere length of the malignant B-cells was not age- dependent. Neither was there a significant correlation between age and the telomere length of the T-cells derived from CLL patients (r ² =0.03, P=0.09; Figure 1 C), suggesting that the T-cell pool does not age in the same way as healthy age-matched individuals. As illustrated in Figures 1 B and 1 C, the CLL B-cells displayed a broad range of telomeres lengths with no apparent lower length threshold. In contrast the mean telomere lengths of the normal T-cell populations from CLL patients displayed a lower length threshold (2.52Kb); T- cells with a mean telomere length less than this threshold could not be detected. This threshold was consistent with that observed in the normal human peripheral blood lymphocytes (PBLs), where mean telomere length of unsorted peripheral blood leukocytes was not detected below 3.10kb (Figure 1 D). This lower length threshold may represent the limit of telomere erosion in normal T-cells having exhausted their replicative capacity.

These results show that the mean telomere length of CLL B-cells and autologous T-cells cannot be inferred from the age of the patient. Normal lymphocytes, including T-cell derived from patients suffering with CLL, display a lower limit to replicative telomere erosion.

CLL B-cell and T-cell telomere length were only weakly correlated but T- cell telomere length was strongly associated with the time since diagnosis

Further evidence for the uncoupling of telomere length in malignant B-cells and autologous T-cells is evidenced by the weak correlation between CLL B-cell telomere length and T-cell telomere length in matched samples from individual patients (r ² =0.076, P=0.045; Figure 2A). When T-cell telomere length was plotted as a function of time from diagnosis to date of sample collection for telomere length analysis, a strong inverse correlation was observed (r ² =0.52, P=0.0002, n=21 ; Figure 2B and r ² =0.13, P= 0.0001 , n=107; Figure 2E). These data suggest that T-cell telomere erosion is an on-going process in CLL and is related to the time since diagnosis. This was confirmed by calculating the average T-cell telomere erosion rate for a subset of samples on which temporally separated samples were available (r ² =0.61 , P=0.06, n=6). In contrast, Figure 2D (r ² =0.09, P=0.18, n=21 ) and Figure 2F (r ² =0.004, P=0.54, n=107) shows that the telomere length of CLL B-cells is not related to the time since diagnosis.

These results show that telomere length of CLL B-cells and autologous T-cells are only weakly correlated. Rather, T-cell telomere length is related to the duration of the disease (time since diagnosis). In contrast, CLL B-cell telomere length was not associated with time since diagnosis.

CD40L co-culture increased Ki-67 expression in CLL B-cells and T-cells We developed an in vitro co-culture system, which allowed patient-derived PBMCs to be maintained for many months. In this way, we were able to investigate telomere dynamics and T-cell function in samples derived from CLL patients. Primary cells were co-cultured on mouse fibroblasts transfected with human CD40 ligand. Co-culture on this‘stromal’ layer led to an increase in the proliferation marker, Ki-67, in the first 28-42 days of the cultures, in both CLL B-cells and T-cells (Figure 3A) and proliferation appears to sustain the viability of cells in the culture (Figure 3B). The importance of T-cells in sustaining these long-term cultures was confirmed by depleting CD3 ⁺ T-cells with magnetic beads. These cultures showed a significant reduction in CLL B-cell Ki-67 expression (Figure 3C) and were unsustainable beyond 28 days pointing to the vital role of autologous T-cells for long-term CLL cell survival even in the presence of the exogenous CD40 ligand presented on the mouse fibroblasts (Figure 3D).

These results show that T-cell and CLL B-cell expression of the proliferation marker, Ki-67, was significantly increased in the co-cultures. Depletion of CD3 ⁺ T-cells resulted in a significant reduction in CLL B-cell proliferation that was associated with the early termination of the cultures. Thus T-cells are required to maintain long-term viability and proliferation of CLL B-cells. Long-term in vitro cultures of primary CLL cells are supported by T-cell cytokine release

Analysis of the supernatants derived from long-term co-cultures, using the MSD V-PLEX pro-inflammatory panel 1 immunoassay kit, confirmed that the T-cells in the long-term culture model secreted a number of pro-inflammatory cytokines, which promote the survival and/or growth of CLL B-cells (Figure 4A). The cellular origin of the cytokines was confirmed in cultures depleted of T-cells, using CD3 beads; supernatants derived from these cultures, contained no detectable IL-2 and significantly reduced levels of INFy, IL-6 and TNFa (Figure 4B). IL-2 levels strongly correlated with Ki-67 expression in the T-cell population and the CLL B-cell population (Figure 4C and 4D respectively).

These results show that pro-inflammatory cytokine expression was significantly increased during the 28-42 days of co-culture. These cytokines were shown to be derived predominantly from autologous T-cells, as evidenced by CD3 depletion experiments. T-cell proliferation was most closely associated with IL-2 production in the co-cultures.

T-cells stop being polyfunctional and fail to support tumour B-cell growth and survival before they reach senescence

T-cells were harvested from co-cultures that had fallen below 5% Ki-67 positive cells (both T-cells and CLL B-cells) and where the total cell count in the culture was less than 5 million. In all cases, the T-cells showed a marked reduction in cytokine production at these time points. T-cells were isolated using negative selection (CD19 beads) and were labelled with CFSE prior to stimulation with CD3/CD28 beads for 3 days. Figure 5 shows that despite the loss of functionality (as evidenced by a marked reduction in cytokine production), and a significant reduction in Ki-67 expression, T-cells derived from these long- term co-cultures were still capable of further cell division following stimulation, so were not senescent. These results show that, despite typically representing less than 5% of the total cell number in the co-cultures, autologous T-cells were the dominant contributors to the pro-inflammatory cytokines in the co-cultures. The decline in T-cell-derived cytokines was mirrored by the decline in CLL cell numbers in the co-cultures. Although the T-cells lost their functionality over the course of the cultures, they were not senescent as they were capable of cell division following stimulation with CD3/CD28 beads.

T-cell telomere length shows significant erosion and an increase in clonality over time in CLL patients

Longitudinal analysis of T-cell telomere length in CLL patients revealed significant telomere erosion over time (Figure 6A). In addition, T-cell populations showed evidence of increased clonality as evidenced by a reduction in standard deviation in the telomere length profiles (Figure 6B). This was further reinforced by T-cell receptor repertoire analysis, which showed further skewing in the TCR repertoire over time (Figure 6C). It is worthy of note that the starting T-cell telomere length in samples added to the in vitro co- culture system was predictive for the duration of the cultures i.e. the time at which the cultures stopped growing (Figure 6D). No culture survived beyond 21 days when the starting mean T-cell telomere length was below 4.34 kb suggesting that this represents an important threshold for sustained T-cell function.

These results show that T-cells derived from CLL patients show significant telomere erosion over the course of the disease, which is accompanied by an increase in clonality in the T-cell receptor repertoire. The T-cell telomere length at the start of the in vitro co-cultures predicts for the duration of proliferation and survival of both T-cells and the leukaemic B-cells. A mean T-cell telomere length <3.47 kb was associated with early termination of the co-culture (within 28 days) suggesting that the functionality of the T-cells was compromised below this length threshold. The telomere length threshold at which functionality is lost, is distinct from that observed in normal T-cells at the limit of their replicative capacity.

Summary

The main findings of this study can be summarised as follows:

Due to a clear relationship between duration of disease and T-cell telomere length, T-cells required for autologous T-cell therapy should be harvested as close to diagnosis as possible so that mean telomere length is not compromised;

Any T-cell based therapeutic should have cells with telomere lengths greater than or equal to a threshold value of 3.47 kb that defines the ability of said cells to sustain a functional response;

Given the well-defined relationship between telomere erosion and cell division in T-cells, the threshold can be used to predict the numbers of population doublings that may be undertaken to generate the required expansion of T-cell based therapeutics without loss of the essential (cytokine release) functionality;

Cytokine secretion by the functional T-cell population is a strong indicator of both T-cell and CLL B-cell proliferation; and

Given that the loss of function precedes senescence, the absence of biomarkers such as CD57, KLRG1 , PD-1 etc., which are associated with senescence, would not be indicative of the suitability of a given population of T-cells for use in a T-cell therapy program. References

1 Park, J. H. et al. Long-Term Follow-up of CD19 CAR Therapy in Acute

Lymphoblastic Leukemia. N Engl J Med 378, 449-459,

doi:10.1056/NEJMoa1709919 (2018).

2 Porter, D. L. et al. Chimeric antigen receptor T cells persist and induce sustained remissions in relapsed refractory chronic lymphocytic leukemia. Science translational medicine 7, 303ra139, doi:10.1 126/scitranslmed.aac5415 (2015).

3 Brudno, J. N. et al. Allogeneic T Cells That Express an Anti-CD19 Chimeric Antigen Receptor Induce Remissions of B-Cell Malignancies That Progress After Allogeneic Hematopoietic Stem-Cell Transplantation Without Causing Graft-Versus-Host Disease. J Clin Oncol 34, 1112-1121 , doi:10.1200/JC0.2015.64.5929 (2016).

4 Ghosh, A. et al. Donor CD19 CAR T cells exert potent graft-versus-lymphoma

activity with diminished graft-versus-host activity. Nat Med 23, 242-249, doi:10.1038/nm.4258 (2017).

5 Harley, C. B., Futcher, A. B. & Greider, C. W. Telomeres shorten during ageing of human fibroblasts. Nature 345, 458-460 (1990).

6 d'Adda di Fagagna, F. et al. A DNA damage checkpoint response in telomere- initiated senescence. Nature 426, 194-198 (2003).

7 Baird, D. M., Rowson, J., Wynford-Thomas, D. & Kipling, D. Extensive allelic

variation and ultrashort telomeres in senescent human cells. Nat Genet 33, 203-207 (2003).

8 Vaziri, H. et al. Loss of telomeric DNA during aging of normal and trisomy 21 human lymphocytes. Am J Hum Genet 52, 661-667 (1993).

9 Bodnar, A. G., Kim, N. W., Effros, R. B. & Chiu, C. P. Mechanism of telomerase induction during T cell activation. Exp Cell Res 228, 58-64 (1996).

10 Roth, A. et al. Telomerase levels control the lifespan of human T lymphocytes. Blood

102, 849-857 (2003).

11 Capper, R. et al. The nature of telomere fusion and a definition of the critical

telomere length in human cells. Genes Dev 21 , 2495-2508 (2007).