Since you can't divide by zero, you can't state the level in the malignant tissue in terms of the expression in normal tissue. You could state it in terms of a housekeeping gene.
The goal is to compare gene expression between malignant and normal tissue, but the gene of interest is not expressed in normal tissue. You could quantify the expression of the gene of interest in a few different ways.
1. You could quantify the expression level of the gene of interest in absolute terms - how many copies of the mRNA are present per cell on average - which would require a synthetic standard in which the number of copies is known.
2. You could compare the expression level of the gene of interest in malignant cells to the expression level of a housekeeping gene in malignant cells. It would also be necessary to compare the expression level of the housekeeping gene in malignant cells to its level in normal cells, in case they differ.
The goal is to give a sense of whether the expression level of the gene of interest in malignant cells is very low, very high, or somewhere in between.
I don't see why you would need a diagram to show this information. It could be given in a table, or even just in the text.