Nine Human Leukocyte Antigen (HLA) Class I Alleles are Omnipotent Against 11 Antigens Expressed in Melanoma Tumors

Abstract

Objective:

Host immunogenetics (Human Leukocyte Antigen, HLA) play a critical role in the human immune response to melanoma, influencing both melanoma prevalence and immunotherapy outcomes. Beneficial outcomes hinge on the successful binding of epitopes of melanoma antigens to HLA Class I molecules for an effective engagement of cytotoxic CD8+ lymphocytes and subsequent elimination of the cancerous cell. This study evaluated the binding affinity and immunogenicity of HLA Class I to melanoma tumor antigens to identify alleles best suited to facilitate elimination of melanoma antigens.

Methods:

In this study, we used freely available software tools to determine in silico the binding affinity and immunogenicity of 2462 reported HLA Class I alleles to all linear nonamer epitopes of 11 known antigens expressed in melanoma tumors (TRP2, S100, Tyrosinase, TRP1, PMEL(17), MAGE1, MAGE4, CTA, BAGE, GAGE/SSX2, Melan).

Results:

We identified the following 9 HLA Class I alleles with very high immunogenicity and binding affinity against all 11 melanoma antigens: A*02:14, B*07:10, B*35:10, B*40:10, B*40:12, B*44:10, C*07:11, and C*07:13, and C*07:14.

Conclusion:

These 9 HLA alleles possess the potential to aid in the elimination of melanoma both by themselves and by enhancing the beneficial effect of immune checkpoint inhibitors.

Keywords

Melanoma neoantigens human leukocyte antigen (HLA)major histocompatibility complex (MHC)immunogenicity

Introduction

Melanoma accounts for a high number of skin cancer related deaths and its incidence has been increasing for decades.^1,2 Therapeutic advances such as immune checkpoint inhibitors (ICI) have improved melanoma survival rates; however, ICI outcomes partially depend on host immunogenetics (Human Leukocyte Antigen, HLA).³ Notably, a recent study documented that the population frequency of HLA alleles associated with favorable ICI outcomes are also negatively associated with the population prevalence of melanoma.⁴ Burgeoning evidence highlighting the influence of HLA on melanoma prevalence, survival, and treatment response suggests a broad influence of HLA on melanoma risk and protection.^3-6

A primary function of HLA involves elimination of foreign antigens including cancer neoantigens.⁷ HLA Class I molecules bind and export melanoma antigen-derived peptides to the cell surface where they are presented to CD8+ T lymphocytes to signal destruction of tumor cells. This process depends on (a) the good binding of tumor epitopes with HLA Class I molecule(s) for the formation of epitope-HLA molecule complex, and (b) the successful engagement by this complex of the T-cell receptor (TCR) for subsequent cytotoxic elimination of the tumor cell. However, due to the highly polymorphic nature of HLA, its molecules vary tremendously with respect to peptide binding capability, with the result that any specific combination of 6 HLA Class I alleles carried by an individual subject may offer only limited coverage of the tumor antigens,^8-16 particularly in light of immune-escape mechanisms exploited by tumors.^17-19 Yet, relatively little is known about binding affinity and immunogenicity of specific HLA alleles with melanoma antigens. Indeed, alleles with high immunogenicity against specific melanoma/cancer antigens would be beneficial in tumor elimination by themselves and by enhancing the effect of immune checkpoint blockade immunotherapy.

In this study we used an in silico approach^20-23 to characterize binding affinity and immunogenicity of 2462 HLA alleles to 11 known melanoma antigens,²⁴ and successfully identified alleles with high immunogenicity and binding affinity against those antigens.

Materials and Methods

Melanoma/cancer antigens

Eleven cancer antigens (Table 1) were used based on their known occurrence in melanoma tumors²⁴; of those 6 are specific to melanoma and 5 are expressed in melanoma as well as in other tumors²⁰ (labeled [CTA] in Table 1). The amino acid (AA) sequences of the 11 melanoma antigens used were retrieved from the Uniprot Database²⁵ and are given in the Appendix.

Table 1.

The 11 cancer/melanoma-related antigens. The listing is from Rodríguez.¹⁹ [CTA] denotes that the antigen is a member of the Cancer Testis Antigen family of antigens also expressed in other tumors.²⁰

	Uniprot ID	Cancer antigen	Protein	N Amino acids
1	O75767	TRP2	Tyrosinase-related protein-2	237
2	P04271	S100	Protein S100-B	92
3	P14679	Tyrosinase	Tyrosinase	529
4	P17643	TRP1	5,6-dihydroxyindole-2-carboxylic acid oxidase	537
5	P40967	PMEL(17)	Melanocyte protein PMEL	661
6	P43355	MAGE1 [CTA]	Melanoma-associated antigen 1	309
7	P43358	MAGE4 [CTA]	Melanoma-associated antigen 4	317
8	P78358	CTA [CTA]	Cancer/testis antigen 1	180
9	Q13072	BAGE [CTA]	B melanoma antigen 1	43
10	Q16385	GAGE, SSX2 [CTA]	Protein SSX2	188
11	Q16655	Melan	Melanoma antigen recognized by T-cells 1	118

In silico determination of immunogenicity of antigen nonamers with HLA Class I alleles

We determined the immunogenicity between 2462 reported HLA Class I alleles²⁶ and 11 melanoma antigens (Table 1) using the INeo-Epp method²² for T-cell receptor (TCR) epitope prediction using theINeo-Epp web tool,²⁰ downloaded in April 2023, and run locally on a CentOS 6.7 server. For that purpose, we split a given melanoma antigen to all possible linear 9-mer (nonamer) AA residue epitopes using a sliding window approach, as detailed previously²⁷ (Figure 1) and submitted each epitope to the application together with a specific HLA allele. More specifically, we paired all epitopes with all alleles and obtained for each pair its percentile rank, a measure of binding affinity of the epitope-HLA allele complex; smaller percentile rank indicates higher binding affinity. The application gave as an outcome a TCR predictive score, $\bar{w}$ , for pairs with high binding affinities (percentile rank < 2); scores > 0.4 indicated positive immunogenicity and were analyzed further (Table 2). We computed the following as a comprehensive measure of immunogenicity for quantitative analyses. Let K be the number of nonamers that showed positive immunogenicity (score > 0.4); then, K weighted by their average score $\bar{w}$ , would serve as a good estimate of the overall effectiveness of a given allele to induce immunogenicity for a given protein:

Total immunogenicity score : TIMG = \bar{w} K

(1)

Figure 1.

Illustration of the sliding nonamer approach used. See text for details.

Table 2.

Number of alleles with allele-antigen immunogenicities above the thresholds mentioned in the text.

Index	UNIPROT	Antigen	N
1	O75767	TRP2	2462
2	P04271	S100	1804
3	P14679	Tyrosinase	2462
4	P17643	TRP1	2462
5	P40967	PMEL(17)	2462
6	P43355	MAGE1	2462
7	P43358	MAGE4	2462
8	P78358	CTA	2252
9	Q13072	BAGE	2112
10	Q16385	GAGE, SSX2	2443
11	Q16655	Melan	2319
		Total	25 702

Finally, TIMG scores were summed for each allele, across all allele-nonamer combinations, and the fractional ranks of these sums were computed for each allele and antigen. For each antigen, the allele with the top-ranked sum (ie, 100%) was the most potent against the particular antigen.

In silico determination of predicted binding affinity of antigen nonamers with HLA Class I alleles

The determination of nonamer immunogenicity above was obtained using the INeo-Epp tool which employs a variety of T-cell immunogenicity-related measures, in addition to the predicted binding affinity of the antigen to the peptide motif of the HLA Class I molecule. As such, the INeo-Epp tool provides a wide coverage of antigen immunogenicity. However, we also determined the predicted binding affinity for the best INeo-Epp outcomes using a different tool, namely the Immune Epitope Database (IEDB) NetMHCpan (ver. 4.1) tool.^21,23 For each nonamer-HLA molecule tested, this tool gives, as an output, the percentile rank of binding affinity of the HLA molecule and the epitope among predicted binding affinities of the same HLA molecule to a large number of different peptides of the same AA length; the smaller the percentile rank, the better the binding affinity. Now, given a protein of N amino acid length and a nonamer, there are N-9 binding affinity predictions, that is, N-9 percentile ranks. Of these predictions, for each melanoma antigen and HLA molecule tested, we retained the lowest percentile rank (LPR) as the best possible binding affinity of the protein-HLA molecule pair.

Results

We identified 2462 reported HLA Class I alleles (Table 2) that possessed high binding affinity to nonamers of 11 known melanoma antigens and yielded HLA-antigen complexes with high immunogenicity (ie, good prediction of engaging cytotoxic CD8+ TCR). We then ranked the alleles according to their immunogenicity strength and identified those with the highest immunogenicity for each melanoma antigen, a total of 9 alleles shown in Table 3. (The results for all alleles are given in Table S1 in Supplemental Material). Remarkably, these 11 alleles were also potent against all other melanoma antigens (Table 4), thus conferring very high protection across the board. The specific nonamers and their highly ranked immunogenicities are shown in Table 5. Finally, Table 6 shows the predicted best binding affinities obtained for the antigen-allele combinations (Table 5) using the NetMHCpan tool. It can be seen that all alleles had excellent predicted binding affinities: all LPR values were <1, a common threshold for a good binding affinity. Thus both INeo-Epp and NetMHCpan prediction tools gave very congruent results.

Table 3.

Alleles with highest immunogenicities (among the 2462 alleles tested) for the different cancer/melanoma antigens.

Index	Protein ID	Cancer antigen	HLA allele
1	O75767	TRP2	C*07:11
2	P04271	S100	B*40:12
3	P14679	Tyrosinase	C*07:13
4	P17643	TRP1	B*35:11
5	P40967	PMEL(17)	A*02:14
6	P43355	MAGE1	A*02:14
7	P43358	MAGE4	B*40:10
8	P78358	CTA	B*07:10
9	Q13072	BAGE	B*44:10
10	Q16385	GAGE, SSX2	C*07:14
11	Q16655	Melan	B*40:12

Table 4.

Fractional ranks of average immunogenicity of alleles in Table 1 across all nonamers. Maximum (best) is 100.

Antigen	A*02:14	B*07:10	B*35:11	B*40:10	B*40:12	B*44:10	C*07:11	C*07:13	C*07:14	Mean
TRP2	99.63	99.76	98.62	98.74	98.38	98.09	100	99.96	99.92	99.23
S100	98.95	59.56	98.67	99.94	100	99.83	70.79	83.20	98.50	89.94
Tyrosinase	99.68	98.42	99.03	98.25	98.17	98.13	99.35	100.0	99.96	99.00
TRP1	99.55	98.86	100	98.33	98.29	98.01	99.84	99.80	99.88	99.17
PMEL(17)	100	99.59	99.23	98.96	98.70	98.21	98.54	98.78	98.66	98.96
MAGE1	100	98.78	99.43	99.07	98.94	98.66	99.51	99.59	99.47	99.27
MAGE4	99.96	98.94	99.19	100	99.92	99.43	98.74	98.78	98.82	99.31
CTA	99.42	100	99.87	98.71	98.62	98.18	98.93	98.89	98.98	99.07
BAGE	98.91	98.39	97.63	99.86	99.91	100	97.77	99.10	99.20	98.97
GAGE, SSX2	98.81	99.59	99.26	99.71	99.63	99.43	99.96	99.92	100	99.59
Melan	98.88	98.66	99.91	99.87	100	99.48	99.96	99.70	99.74	99.58

Values in bold indicate the highest predicted affinities for a given antigen (row) across the 9 HLA alleles (columns).

Table 5.

Nonamers of best allele for each melanoma antigen. Begin and End denote the beginning and end of the nonamer within the antigen AA sequence (Table S3).

Antigen Index	UNIPROT	Antigen	Nonamer index	Nonamer	Begin	End	Best allele
1	O75767	TRP2	1	SPLWWGFLL	2	11	C07:11
1	O75767	TRP2	2	QFPRVCMTV	24	33	C07:11
1	O75767	TRP2	3	TRPWSGPYI	68	77	C07:11
1	O75767	TRP2	4	LRNQDDREL	77	86	C07:11
1	O75767	TRP2	5	SPQEREQFL	134	143	C07:11
1	O75767	TRP2	6	EREQFLGAL	137	146	C07:11
1	O75767	TRP2	7	YVITTQHWV	156	165	C07:11
1	O75767	TRP2	8	SVYDFFVWL	180	189	C07:11
1	O75767	TRP2	9	HYYSVRDTL	189	198	C07:11
1	O75767	TRP2	10	YYSVRDTLL	190	199	C07:11
1	O75767	TRP2	11	VRDTLLGGF	193	202	C07:11
1	O75767	TRP2	12	YRFVIGLRV	210	219	C07:11

2	P04271	S100	1	SELEKAMVA	2	11	B40:12
2	P04271	S100	2	LEKAMVALI	4	13	B40:12
2	P04271	S100	3	AMVALIDVF	7	16	B40:12
2	P04271	S100	4	KEQEVVDKV	49	58	B40:12
2	P04271	S100	5	GECDFQEFM	67	76	B40:12
2	P04271	S100	6	QEFMAFVAM	72	81	B40:12

3	P14679	TYROSINASE	1	SFQTSAGHF	12	21	C07:13
3	P14679	TYROSINASE	2	SNAPLGPQF	61	70	C07:13
3	P14679	TYROSINASE	3	YLTLAKHTI	137	146	C07:13
3	P14679	TYROSINASE	4	FAHEAPAFL	200	209	C07:13
3	P14679	TYROSINASE	5	AFLPWHRLF	206	215	C07:13
3	P14679	TYROSINASE	6	FLPWHRLFL	207	216	C07:13
3	P14679	TYROSINASE	7	GQHPTNPNL	254	263	C07:13
3	P14679	TYROSINASE	8	QHPTNPNLL	255	264	C07:13
3	P14679	TYROSINASE	9	GSMDKAANF	330	339	C07:13
3	P14679	TYROSINASE	10	SFRNTLEGF	339	348	C07:13
3	P14679	TYROSINASE	11	GFASPLTGI	346	355	C07:13
3	P14679	TYROSINASE	12	SQSSMHNAL	358	367	C07:13
3	P14679	TYROSINASE	13	SSMHNALHI	360	369	C07:13
3	P14679	TYROSINASE	14	YMNGTMSQV	369	378	C07:13
3	P14679	TYROSINASE	15	IFLLHHAFV	385	394	C07:13
3	P14679	TYROSINASE	16	RRHRPLQEV	402	411	C07:13
3	P14679	TYROSINASE	17	VYPEANAPI	410	419	C07:13
3	P14679	TYROSINASE	18	SYMVPFIPL	424	433	C07:13
3	P14679	TYROSINASE	19	YMVPFIPLY	425	434	C07:13
3	P14679	TYROSINASE	20	SYLEQASRI	466	475	C07:13
3	P14679	TYROSINASE	21	MVGAVLTAL	483	492	C07:13
3	P14679	TYROSINASE	22	ALLAGLVSL	490	499	C07:13

4	P17643	TRP1	1	RPHSPQYPH	73	82	B35:11
4	P17643	TRP1	2	GPDGNTPQF	171	180	B35:11
4	P17643	TRP1	3	FLGVGQESF	200	209	B35:11
4	P17643	TRP1	4	DFSHEGPAF	212	221	B35:11
4	P17643	TRP1	5	FSHEGPAFL	213	222	B35:11
4	P17643	TRP1	6	DLMGSRSNF	264	273	B35:11
4	P17643	TRP1	7	VARPMVQRL	319	328	B35:11
4	P17643	TRP1	8	EPQDVAQCL	329	338	B35:11
4	P17643	TRP1	9	GLFDTPPFY	340	349	B35:11
4	P17643	TRP1	10	FYSNSTNSF	347	356	B35:11
4	P17643	TRP1	11	SLHNLAHLF	375	384	B35:11
4	P17643	TRP1	12	FPLENAPIG	425	434	B35:11
4	P17643	TRP1	13	VPFWPPVTN	441	450	B35:11
4	P17643	TRP1	14	WPPVTNTEM	444	453	B35:11
4	P17643	TRP1	15	PPVTNTEMF	445	454	B35:11
4	P17643	TRP1	16	VTAPDNLGY	454	463	B35:11
4	P17643	TRP1	17	EIQWPSREF	465	474	B35:11
4	P17643	TRP1	18	VPEIIAIAV	475	484	B35:11
4	P17643	TRP1	19	IAIAVVGAL	479	488	B35:11
4	P17643	TRP1	20	IAVVGALLL	481	490	B35:11
4	P17643	TRP1	21	YLIRARRSM	499	508	B35:11

5	P40967	PMEL(17)	1	HLAVIGALL	11	20	A02:14
5	P40967	PMEL(17)	2	AVIGALLAV	13	22	A02:14
5	P40967	PMEL(17)	3	LLAVGATKV	18	27	A02:14
5	P40967	PMEL(17)	4	QLYPEWTEA	47	56	A02:14
5	P40967	PMEL(17)	5	KVSNDGPTL	69	78	A02:14
5	P40967	PMEL(17)	6	TLIGANASF	76	85	A02:14
5	P40967	PMEL(17)	7	KVLPDGQVI	95	104	A02:14
5	P40967	PMEL(17)	8	KTWGQYWQV	154	163	A02:14
5	P40967	PMEL(17)	9	VLGGPVSGL	162	171	A02:14
5	P40967	PMEL(17)	10	RAMLGTHTM	176	185	A02:14
5	P40967	PMEL(17)	11	GTHTMEVTV	180	189	A02:14
5	P40967	PMEL(17)	12	ITDQVPFSV	209	218	A02:14
5	P40967	PMEL(17)	13	FLRNQPLTF	232	241	A02:14
5	P40967	PMEL(17)	14	QLHDPSGYL	243	252	A02:14
5	P40967	PMEL(17)	15	ALVVTHTYL	273	282	A02:14
5	P40967	PMEL(17)	16	YLEPGPVTA	280	289	A02:14
5	P40967	PMEL(17)	17	AQVVLQAAI	288	297	A02:14
5	P40967	PMEL(17)	18	VVLQAAIPL	290	299	A02:14
5	P40967	PMEL(17)	19	VLQAAIPLT	291	300	A02:14
5	P40967	PMEL(17)	20	GQVPTTEVV	325	334	A02:14
5	P40967	PMEL(17)	21	SVQVPTTEV	350	359	A02:14
5	P40967	PMEL(17)	22	VQVPTTEVI	351	360	A02:14
5	P40967	PMEL(17)	23	GMTPEKVPV	373	382	A02:14
5	P40967	PMEL(17)	24	GMTPAEVSI	399	408	A02:14
5	P40967	PMEL(17)	25	VVLSGTTAA	408	417	A02:14
5	P40967	PMEL(17)	26	AQVTTTEWV	416	425	A02:14
5	P40967	PMEL(17)	27	ITGSLGPLL	450	459	A02:14
5	P40967	PMEL(17)	28	PLLDGTATL	456	465	A02:14
5	P40967	PMEL(17)	29	RQVPLDCVL	469	478	A02:14
5	P40967	PMEL(17)	30	ESAEILQAV	494	503	A02:14
5	P40967	PMEL(17)	31	VLPSPACQL	544	553	A02:14
5	P40967	PMEL(17)	32	SLAVVSTQL	576	585	A02:14
5	P40967	PMEL(17)	33	IMPGQEAGL	585	594	A02:14
5	P40967	PMEL(17)	34	GQEAGLGQV	588	597	A02:14
5	P40967	PMEL(17)	35	AGLGQVPLI	591	600	A02:14
5	P40967	PMEL(17)	36	GLGQVPLIV	592	601	A02:14
5	P40967	PMEL(17)	37	GILLVLMAV	601	610	A02:14
5	P40967	PMEL(17)	38	LMAVVLASL	606	615	A02:14
5	P40967	PMEL(17)	39	KQDFSVPQL	622	631	A02:14

6	P43355	MAGE1	1	AQQEALGLV	18	27	A02:14
6	P43355	MAGE1	2	ATSSSSPLV	31	40	A02:14
6	P43355	MAGE1	3	LVLGTLEEV	38	47	A02:14
6	P43355	MAGE1	4	GTLEEVPTA	41	50	A02:14
6	P43355	MAGE1	5	SAFPTTINF	62	71	A02:14
6	P43355	MAGE1	6	KVADLVGFL	105	114	A02:14
6	P43355	MAGE1	7	VADLVGFLL	106	115	A02:14
6	P43355	MAGE1	8	KASESLQLV	146	155	A02:14
6	P43355	MAGE1	9	GLLGDNQIM	181	190	A02:14
6	P43355	MAGE1	10	FLWGPRALA	264	273	A02:14
6	P43355	MAGE1	11	RALAETSYV	269	278	A02:14
6	P43355	MAGE1	12	RVRFFFPSL	289	298	A02:14
6	P43355	MAGE1	13	FFFPSLREA	292	301	A02:14
6	P43355	MAGE1	14	ALREEEEGV	301	310	A02:14

7	P43358	MAGE4	1	AQEEALGLV	18	27	B40:10
7	P43358	MAGE4	2	GASALPTTI	68	77	B40:10
7	P43358	MAGE4	3	KVDELAHFL	113	122	B40:10
7	P43358	MAGE4	4	VDELAHFLL	114	123	B40:10
7	P43358	MAGE4	5	KELVTKAEM	128	137	B40:10
7	P43358	MAGE4	6	SEEEIWEEL	218	227	B40:10
7	P43358	MAGE4	7	EEIWEELGV	220	229	B40:10
7	P43358	MAGE4	8	WEELGVMGV	223	232	B40:10
7	P43358	MAGE4	9	RQVPGSNPA	260	269	B40:10
7	P43358	MAGE4	10	YEFLWGPRA	270	279	B40:10

8	P78358	CTA	1	ARGPESRLL	80	89	B07:10
8	P78358	CTA	2	LAMPFATPM	92	101	B07:10
8	P78358	CTA	3	MPFATPMEA	94	103	B07:10
8	P78358	CTA	4	FATPMEAEL	96	105	B07:10
8	P78358	CTA	5	SLAQDAPPL	108	117	B07:10
8	P78358	CTA	6	AQDAPPLPV	110	119	B07:10
8	P78358	CTA	7	APPLPVPGV	113	122	B07:10
8	P78358	CTA	8	PPLPVPGVL	114	123	B07:10
8	P78358	CTA	9	AADHRQLQL	139	148	B07:10

9	Q13072	BAGE	1	AQLLQARLM	12	21	B44:10
9	Q13072	BAGE	2	KEESPVVSW	21	30	B44:10
9	Q13072	BAGE	3	LEPEDGTAL	31	40	B44:10
9	Q13072	BAGE	4	PEDGTALCF	33	42	B44:10

10	Q16385	GAGE,SSX	1	MTFGRLQGI	1	10	C07:14
10	Q16385	GAGE,SSX	2	DAFARRPTV	5	14	C07:14
10	Q16385	GAGE,SSX	3	RRPTVGAQI	9	18	C07:14
10	Q16385	GAGE,SSX	4	FKATLPPFM	63	72	C07:14

11	Q16655	MELAN	1	AEEAAGIGI	24	33	B40:12
11	Q16655	MELAN	2	EEAAGIGIL	25	34	B40:12
11	Q16655	MELAN	3	RRNGYRALM	50	59	B40:12

Table 6.

Predicted best binding affinities (LPR values) using the NetChan tool. The lower the LPR, the higher the predicted affinity.

Antigen	A*02:14	B*07:10	B*35:11	B*40:10	B*40:12	B*44:10	C*07:11	C*07:13	C*07:14	Mean
TRP2	0.05	0.1	0.04	0.12	0.11	0.11	0.03	0.02	0.05	0.070
S100	0.5	0.92	0.12	0.05	0.1	0.05	0.44	0.89	0.55	0.402
Tyrosinase	0.07	0.02	0.04	0.03	0.02	0.02	0.02	0.02	0.03	0.030
TRP1	0.02	0.03	0.06	0.04	0.02	0.06	0.03	0.02	0.02	0.033
PMEL(17)	0.03	0.03	0.03	0.03	0.01	0.01	0.04	0.04	0.12	0.038
MAGE1	0.03	0.06	0.05	0.07	0.14	0.04	0.16	0.31	0.08	0.104
MAGE4	0.03	0.04	0.04	0.03	0.06	0.04	0.1	0.18	0.23	0.083
CTA	0.08	0.07	0.01	0.07	0.05	0.02	0.15	0.2	0.17	0.091
BAGE	0.44	0.14	0.91	0.32	0.52	0.11	0.12	0.32	0.38	0.362
GAGE, SSX2	0.04	0.04	0.2	0.09	0.15	0.06	0.03	0.1	0.12	0.092
Melan	0.05	0.03	0.1	0.18	0.39	0.1	0.03	0.03	0.02	0.103

Values in bold indicate the highest predicted affinities for a given antigen (row) across the 9 HLA alleles (columns).

Discussion

Here we identified in silico 9 HLA Class I alleles with the highest immunogenicity (out of 2462 tested) against 11 melanoma antigens. These alleles, and others with lower but still high immunogenicities against the tested antigens (See Table 1 in Supplemental Material), are excellent candidates as therapeutic anti-melanoma agents, acting both by themselves (inducing the lysis of cancerous cells via engagement of cytotoxic CD8+ T-cells) and by aiding ICI, the success of which has been found to depend on the HLA Class I genetic makeup of the patient.³

Since HLA is the most highly polymorphic region of the human genome and a particular individual carries only 6 Class I HLA alleles, the likelihood that a given individual may possess a HLA allele that can bind melanoma antigens with sufficient immunogenicity to be substantially effective in tumor cell elimination is limited. Thus, the outcome of ICI would be substantially potentiated if such HLA Class I molecules were present in the tumor cell. Indeed, we have recently proposed²⁸ that this could be achieved by inducing the biosynthesis of desired HLA Class I molecules in the tumor cells by injecting in the tumor site the suitably LNP-mRNA (i.e. mRNA packaged with lipid nanoparticles, LNP) of those Class I molecules, (“Direct HLA therapy, dHLA”²⁸). In the specific case of melanoma, the treatment would involve the introduction in the tumor site of the LNP-mRNA of the 9 omnipotent alleles in Table 3 using recently improved methods of administration.²⁹ An additional benefit could be derived indirectly from the well-known immunogenicity of the HLA molecules themselves, that is, the production of antibodies against them leading, for example, to the rejection of a transplanted organ. In the present application, such an effect would actually be beneficial, since it will enhance the elimination of the tumor, which would be rejected as a foreign organ. This treatment could be applied to all forms of solid tumors for which tumor antigens and suitably potent HLA Class I alleles against them have been identified, a feasible objective, as demonstrated in the present study. In this context, it is noteworthy that 5 of the 11 antigens tested, namely the antigens in the Cancer Testis Antigen (CTA) family (Table 3) are also expressed in other tumors^24,30-32 (e.g. lung carcinoma,³⁰ head and neck squamous cell carcinoma³¹) and, therefore, administration of the 9 alleles discovered here (Tables 4 and 6) would be beneficial to such tumors, in addition to melanoma.

Finally, it is important to distinguish the proposed therapy here from cancer vaccine therapies currently being developed and applied.^33-36 In a cancer vaccine, cancer (neo)antigens are suitably administered to induce the production of antibodies against them and thus reduction/elimination of the tumor, engaging antigen presenting cells (APC), the HLA Class II system, CD4+ lymphocytes, B cells, etc. In contrast, in our proposed dHLA application,²⁸ the mRNA of specific HLA Class I molecules, found to bind with high affinity to cancer antigen fragments (nonamers), are injected directly into the tumor, where the protein of the HLA molecule is synthesized, binds to antigen nonamers, is transported to the cell surface, and engages the cytotoxic CD8+ lymphocytes, killing the cancerous cell. Thus this method does not involve the production of antibodies but depends on the ability of cytotoxic lymphocytes to kill the cancer cell, an ability that, when reduced or absent, can be restored by ICI, hence the prospect of dHLA therapy as a promising aid in ICI success, in addition to its expected effectiveness in early stages of cancer when immune blockade is weak or absent. Given the cost and toxicity of current cancer immunotherapies,^37-39 dHLA therapy²⁸ would be a good addition: the cost should be low (just making the LNP-mRNA of the 9 HLA molecules) and adverse effects minimal, if any, given the local, intratumoral injection of natural molecules already present in humans. In a way, dHLA cancer therapy opens a new avenue in response to the “need for innovation” in cancer immunotherapy requested recently.⁴⁰

Limitations

The main limitation is that these effective alleles are to be injected to tumors locally, and not systemically. Thus, they are not suitable for diffuse tumors, multiple metastases, or inaccessible sites.

Conclusions

The 9 HLA alleles identified here to possess high affinity and immunogenicity against 11 common cancer antigens are specific, key molecules for use in direct HLA cancer therapy with expected favorable application to melanoma and other solid tumors, and minimal (if any) adverse effects. Their evaluation in preclinical assessments and clinical trials regarding tumor reduction/elimination remains to be determined with respect to effects of themselves alone and/or administered in combination with ICI.

Supplemental Material

sj-docx-1-cix-10.1177_11769351241274160 – Supplemental material for Nine Human Leukocyte Antigen (HLA) Class I Alleles are Omnipotent Against 11 Antigens Expressed in Melanoma Tumors

Supplemental material, sj-docx-1-cix-10.1177_11769351241274160 for Nine Human Leukocyte Antigen (HLA) Class I Alleles are Omnipotent Against 11 Antigens Expressed in Melanoma Tumors by Apostolos P Georgopoulos, Lisa M James and Matthew Sanders in Cancer Informatics

Footnotes

Appendix

Amino acid sequences of the 11 melanoma antigens analyzed (Table 1).

MSPLWWGFLLSCLGCKILPGAQGQFPRVCMTVDSLVNKECCPRLGAESANVCGSQQGRGQ

CTEVRADTRPWSGPYILRNQDDRELWPRKFFHRTCKCTGNFAGYNCGDCKFGWTGPNCER

KKPPVIRQNIHSLSPQEREQFLGALDLAKKRVHPDYVITTQHWVGLLGPNGTQPQFANCS

VYDFFVWLHYYSVRDTLLGGFFPWLKVYYYRFVIGLRVWQWEVISCKLIKRATTRQP

MSELEKAMVALIDVFHQYSGREGDKHKLKKSELKELINNELSHFLEEIKEQEVVDKVMET

LDNDGDGECDFQEFMAFVAMVTTACHEFFEHE

MLLAVLYCLLWSFQTSAGHFPRACVSSKNLMEKECCPPWSGDRSPCGQLSGRGSCQNILL

SNAPLGPQFPFTGVDDRESWPSVFYNRTCQCSGNFMGFNCGNCKFGFWGPNCTERRLLVR

RNIFDLSAPEKDKFFAYLTLAKHTISSDYVIPIGTYGQMKNGSTPMFNDINIYDLFVWMH

YYVSMDALLGGSEIWRDIDFAHEAPAFLPWHRLFLLRWEQEIQKLTGDENFTIPYWDWRD

AEKCDICTDEYMGGQHPTNPNLLSPASFFSSWQIVCSRLEEYNSHQSLCNGTPEGPLRRN

PGNHDKSRTPRLPSSADVEFCLSLTQYESGSMDKAANFSFRNTLEGFASPLTGIADASQS

SMHNALHIYMNGTMSQVQGSANDPIFLLHHAFVDSIFEQWLRRHRPLQEVYPEANAPIGH

NRESYMVPFIPLYRNGDFFISSKDLGYDYSYLQDSDPDSFQDYIKSYLEQASRIWSWLLG

AAMVGAVLTALLAGLVSLLCRHKRKQLPEEKQPLLMEKEDYHSLYQSHL

MSAPKLLSLGCIFFPLLLFQQARAQFPRQCATVEALRSGMCCPDLSPVSGPGTDRCGSSS

GRGRCEAVTADSRPHSPQYPHDGRDDREVWPLRFFNRTCHCNGNFSGHNCGTCRPGWRGA

ACDQRVLIVRRNLLDLSKEEKNHFVRALDMAKRTTHPLFVIATRRSEEILGPDGNTPQFE

NISIYNYFVWTHYYSVKKTFLGVGQESFGEVDFSHEGPAFLTWHRYHLLRLEKDMQEMLQ

EPSFSLPYWNFATGKNVCDICTDDLMGSRSNFDSTLISPNSVFSQWRVVCDSLEDYDTLG

TLCNSTEDGPIRRNPAGNVARPMVQRLPEPQDVAQCLEVGLFDTPPFYSNSTNSFRNTVE

GYSDPTGKYDPAVRSLHNLAHLFLNGTGGQTHLSPNDPIFVLLHTFTDAVFDEWLRRYNA

DISTFPLENAPIGHNRQYNMVPFWPPVTNTEMFVTAPDNLGYTYEIQWPSREFSVPEIIA

IAVVGALLLVALIFGTASYLIRARRSMDEANQPLLTDQYQCYAEEYEKLQNPNQSVV

MDLVLKRCLLHLAVIGALLAVGATKVPRNQDWLGVSRQLRTKAWNRQLYPEWTEAQRLDC

WRGGQVSLKVSNDGPTLIGANASFSIALNFPGSQKVLPDGQVIWVNNTIINGSQVWGGQP

VYPQETDDACIFPDGGPCPSGSWSQKRSFVYVWKTWGQYWQVLGGPVSGLSIGTGRAMLG

THTMEVTVYHRRGSRSYVPLAHSSSAFTITDQVPFSVSVSQLRALDGGNKHFLRNQPLTF

ALQLHDPSGYLAEADLSYTWDFGDSSGTLISRALVVTHTYLEPGPVTAQVVLQAAIPLTS

CGSSPVPGTTDGHRPTAEAPNTTAGQVPTTEVVGTTPGQAPTAEPSGTTSVQVPTTEVIS

TAPVQMPTAESTGMTPEKVPVSEVMGTTLAEMSTPEATGMTPAEVSIVVLSGTTAAQVTT

TEWVETTARELPIPEPEGPDASSIMSTESITGSLGPLLDGTATLRLVKRQVPLDCVLYRY

GSFSVTLDIVQGIESAEILQAVPSGEGDAFELTVSCQGGLPKEACMEISSPGCQPPAQRL

CQPVLPSPACQLVLHQILKGGSGTYCLNVSLADTNSLAVVSTQLIMPGQEAGLGQVPLIV

GILLVLMAVVLASLIYRRRLMKQDFSVPQLPHSSSHWLRLPRIFCSCPIGENSPLLSGQQ

MSLEQRSLHCKPEEALEAQQEALGLVCVQAATSSSSPLVLGTLEEVPTAGSTDPPQSPQG

ASAFPTTINFTRQRQPSEGSSSREEEGPSTSCILESLFRAVITKKVADLVGFLLLKYRAR

EPVTKAEMLESVIKNYKHCFPEIFGKASESLQLVFGIDVKEADPTGHSYVLVTCLGLSYD

GLLGDNQIMPKTGFLIIVLVMIAMEGGHAPEEEIWEELSVMEVYDGREHSAYGEPRKLLT

QDLVQEKYLEYRQVPDSDPARYEFLWGPRALAETSYVKVLEYVIKVSARVRFFFPSLREA

ALREEEEGV

MSSEQKSQHCKPEEGVEAQEEALGLVGAQAPTTEEQEAAVSSSSPLVPGTLEEVPAAESA

GPPQSPQGASALPTTISFTCWRQPNEGSSSQEEEGPSTSPDAESLFREALSNKVDELAHF

LLRKYRAKELVTKAEMLERVIKNYKRCFPVIFGKASESLKMIFGIDVKEVDPASNTYTLV

TCLGLSYDGLLGNNQIFPKTGLLIIVLGTIAMEGDSASEEEIWEELGVMGVYDGREHTVY

GEPRKLLTQDWVQENYLEYRQVPGSNPARYEFLWGPRALAETSYVKVLEHVVRVNARVRI

AYPSLREAALLEEEEGV

MQAEGRGTGGSTGDADGPGGPGIPDGPGGNAGGPGEAGATGGRGPRGAGAARASGPGGGA

PRGPHGGAASGLNGCCRCGARGPESRLLEFYLAMPFATPMEAELARRSLAQDAPPLPVPG

VLLKEFTVSGNILTIRLTAADHRQLQLSISSCLQQLSLLMWITQCFLPVFLAQPPSGQRR

MAARAVFLALSAQLLQARLMKEESPVVSWRLEPEDGTALCFIF

MNGDDAFARRPTVGAQIPEKIQKAFDDIAKYFSKEEWEKMKASEKIFYVYMKRKYEAMTK

LGFKATLPPFMCNKRAEDFQGNDLDNDPNRGNQVERPQMTFGRLQGISPKIMPKKPAEEG

NDSEEVPEASGPQNDGKELCPPGKPTTSEKIHERSGPKRGEHAWTHRLRERKQLVIYEEI

SDPEEDDE

MPREDAHFIYGYPKKGHGHSYTTAEEAAGIGILTVILGVLLLIGCWYCRRRNGYRALMDK

SLHVGTQCALTRRCPQEGFDHRDSKVSLQEKNCEPVVPNAPPAYEKLSAEQSPPPYSP

Acknowledgements

None.

Author Contributions

APG conceived the proposal and analyzed the data. MS performed the INeo-Epp and NetChan analyses. APG and LMJ wrote the paper. APG, LMJ and MS reviewed and approved the paper.

Availability of Data and Materials

All data used for analysis are freely available from public databases, as stated in the Materials and Methods Section of the manuscript.

Declaration Of Conflicting Interests:

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding:

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Partial funding for this study was provided by the University of Minnesota (the Kunin Chair in Women’s Healthy Brain Aging, the Brain and Genomics Fund, and the American Legion Brain Sciences Chair) and the U.S. Department of Veterans Affairs. The sponsors had no role in the current study design, analysis or interpretation, or in the writing of this paper. The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

Ethical Approval and Consent to Participate

This article simply and solely analyzed publicly available data and does not contain any studies with human participants or animal experiments performed by any of the authors. Hence, ethical approval is not applicable to this study.

Consent for Publication

Not applicable.

ORCID iD

Apostolos P Georgopoulos

Supplemental Material

Supplemental material for this article is available online.

References

National Cancer Institute. Melanoma of the skin-cancer stat facts. Accessed January 12, 2023. https://seer.cancer.gov/statfacts/html/melan.html

Sung

Ferlay

Siegel

, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71:209-249.

Chowell

Morris

Grigg

, et al. Patient HLA class I genotype influences cancer response to checkpoint blockade immunotherapy. Science. 2017;359:582-587.

James

Georgopoulos

. Protective and susceptibility effects of human leukocyte antigen on melanoma prevalence and their implications for predicting checkpoint blockade immunotherapy outcomes. J Immunol Sci. 2022;6:25-35.

Mendez

Aptsiauri

Del Campo

, et al. HLA and melanoma: multiple alterations in HLA class I and II expression in human melanoma cell lines from ESTDAB cell bank. Cancer Immunol Immunother. 2009;58:1507-1515.

Dhall

Patiyal

Kaur

, et al. Computing skin cutaneous melanoma outcome from the HLA-alleles and clinical characteristics. Front Genet. 2020;11:221.

Zhang

Qin

, et al. Neoantigen: a new breakthrough in tumor immunotherapy. Front Immunol. 2021;12:672356.

Hov

Kosmoliaptsis

Traherne

, et al. Electrostatic modifications of the human leukocyte antigen-DR P9 peptide-binding pocket and susceptibility to primary sclerosing cholangitis. Hepatology. 2011;53:1967-1976.

Davenport

Quinn

Chicz

, et al. Naturally processed peptides from two disease-resistance-associated HLA-DR13 alleles show related sequence motifs and the effects of the dimorphism at position 86 of the HLA-DR beta chain. Proc Natl Acad Sci USA. 1995;92:6567-6571.

10.

Charonis

James

Georgopoulos

. In silico assessment of binding affinities of three dementia-protective human leukocyte antigen (HLA) alleles to nine human herpes virus antigens. Curr Res Transl Med. 2020;68:211-216.

11.

Charonis

Tsilibary

Georgopoulos

. SARS-CoV-2 virus and human leukocyte antigen (HLA) class II: investigation in silico of binding affinities for COVID-19 protection and vaccine development. J Immunol Sci. 2020;4:12-23.

12.

Charonis

Tsilibary

E-P

Georgopoulos

. In silico investigation of binding affinities between human leukocyte antigen class I molecules and SARS-CoV-2 virus spike and ORF1ab proteins. Explor Immunol. 2021;1:16-26.

13.

Charonis

James

Georgopoulos

. SARS-CoV-2 in silico binding affinity to human leukocyte antigen (HLA) Class II molecules predicts vaccine effectiveness across variants of concern (VOC). Sci Rep. 2022;12:8074.

14.

Dhanda

Karosiene

Edwards

, et al. Predicting HLA CD4 immunogenicity in human populations. Front Immunol. 2018;9:1369.

15.

Lucas

Leffell

Zachary

. Differences in immunogenicity of HLA antigens and the impact of cross-reactivity on the humoral response. Transplantation. 2015;99:77-85.

16.

Dendrou

Petersen

Rossjohn

Fugger

. HLA variation and disease. Nat Rev Immunol. 2018;18:325-339.

17.

Hazini

Fisher

Seymour

. Deregulation of HLA-I in cancer and its central importance for immunotherapy. J Immunother Cancer. 2021;9:e002899.

18.

Wang

Xiong

Hsu

Wang

Chen

. Human leukocyte antigen (HLA) and cancer immunotherapy: HLA-dependent and -independent adoptive immunotherapies. Ann Blood. 2020;5:14-70.

19.

Rodríguez

. HLA-mediated tumor escape mechanisms that may impair immunotherapy clinical outcomes via T-cell activation. Oncol Lett. 2017;14:4415-4427.

20.

INeo-Epp: a novel T-cell HLA Class-I immunogenicity prediction method. Accessed September 21, 2022. http://www.biostatistics.online/ineo-epp/neoantigen.php

21.

Immune Epitope Database and Tools. IEDB Analysis Resource. Accessed April 11, 2024. http://tools.iedb.org/mhci/

22.

Wang

Wan

Jian

, et al. INeo-Epp: a novel T-cell HLA class-I immunogenicity or neoantigenic epitope prediction method based on sequence-related amino acid features. Biomed Res Int. 2020;2020:5798356. doi: 10.1155/2020/5798356.

23.

Reynisson

Alvarez

Paul

Peters

Nielsen

.NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data. Nucleic Acids Res. 2020;48:W449-W454.

24.

Pitcovski

Shahar

Aizenshtein

Gorodetsky

. Melanoma antigens and related immunological markers. Crit Rev Oncol Hematol. 2017;115:36-49.

25.

The UniProt Consortium. UniProt: the universal protein knowledgebase. Nucleic Acids Res. 2017;45:D158-D169.

26.

Gonzalez-Galarza

McCabe

Santos

, et al. Allele frequency net database. Accessed September 12, 2022. https://www.allelefrequencies.net/

27.

Georgopoulos

James

Charonis

Sanders

. Melanoma and human leukocyte antigen (HLA): Immunogenicity of 69 HLA Class I alleles with 11 antigens expressed in melanoma tumors. Cancer Inform. 2023;22:1172604. doi:10.1177/11769351231172604.

28.

Georgopoulos

James

. Direct administration of human leucocyte antigen (dHLA) molecules into tumour sites: proposal for a new immunotherapy for cancer. BJC Reports. 2023;1:17.

29.

Muñoz

Williams

Dixon

, et al. Influence of injection technique, drug formulation and tumor microenvironment on intratumoral immunotherapy delivery and efficacy. J Immunother Cancer. 2021;9:e001800. doi:10.1136/jitc-2020-001800.

30.

Gjerstorff

Pøhl

Olsen

Ditzel

. Analysis of GAGE, NY-ESO-1 and SP17 cancer/testis antigen expression in early stage non-small cell lung carcinoma. BMC Cancer. 2013;13:466.

31.

Piotti

Scognamiglio

Chiu

Chen

. Expression of cancer/testis (CT) antigens in squamous cell carcinoma of the head and neck: evaluation as markers of squamous dysplasia. Pathol Res Pract. 2013;209:721-726.

32.

Caballero

Chen

. Cancer/testis (CT) antigens: potential targets for immunotherapy. Cancer Sci. 2009;100:2014-2021.

33.

Saxena

van der Burg

Melief

CJM

Bhardwaj

. Therapeutic cancer vaccines. Nat Rev Cancer. 2021;21:360-378.

34.

Igarashi

Sasada

. Cancer vaccines: toward the next breakthrough in cancer immunotherapy. J Immunol Res. 2020;2020:13.

35.

Liu

Huang

Yang

Deng

. mRNA cancer vaccines: construction and boosting strategies. ACS Nano. 2023;17:19550-19580.

36.

Morse

Gwin

Mitchell

. Vaccine therapies for cancer: then and now. Target Oncol. 2021;16:121-152.

37.

Kennedy

Salama

AKS

. A review of cancer immunotherapy toxicity. CA Cancer J Clin. 2020;70:86-104.

38.

Chhabra

Kennedy

. A review of cancer immunotherapy toxicity: immune checkpoint inhibitors. J Med Toxicol. 2021;17:411-424.

39.

Chhabra

Kennedy

. A review of cancer immunotherapy toxicity ii: adoptive cellular therapies, kinase inhibitors, monoclonal antibodies, and oncolytic viruses. J Med Toxicol. 2022;18:43-55.

40.

Johnson

Gainor

Sullivan

Longo

Chabner

. Immune checkpoint inhibitors — the need for innovation. New Engl J Med. 2023;388:1529-1532.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.34 MB